Warn when Roots, Sampling, or Logging APIs are used with MCP protocol version 2026-07-28 or newer.
Add a shared protocol deprecation helper, YARD deprecation notes, and regression tests covering warnings at and after the deprecation threshold and no warnings for older or never-established protocol versions.
Closes https://github.com/modelcontextprotocol/ruby-sdk/issues/390
Co-authored-by: Qlaw <noreply@qlaw.quick.shopify.io>
Client#tools/#list_tools built Tool objects without an annotations
field even though the server includes annotations in every
tools/list response. Add the reader and pass the raw hash through,
mirroring how output_schema is already handled.
## Motivation and Context
`MCP::Client::HTTP` buffers server bytes with no upper bound in two places:
- SSE responses are parsed incrementally, but the parser accumulates `data:` lines until
a blank line dispatches the event, so a server that never terminates an event grows
the buffer indefinitely. The abort-on-response signal does not help here, since it only fires
after a complete response event has been dispatched.
- Non-SSE (JSON) bodies accumulate in `SSEStream#buffer` chunk by chunk with no limit.
Both the POST response stream and the standalone GET listening stream are affected.
The stdio client already bounds server frames (`max_line_bytes:`, 4 MiB default)
and the server transports bound request bodies (`max_request_bytes:`, 4 MiB default);
this brings the HTTP client in line.
`MCP::Client::HTTP.new` gains `max_message_bytes:` (default 4 MiB), which caps the bytes buffered
for a single message - an SSE event or a JSON response body. External byte counting cannot tell
consumed-and-discarded bytes (comments, field names, delimiters) from consumed-and-retained ones,
so the check measures the parser's own String buffers after each chunk. An over-limit message raises
`RequestHandlerError` on the request path and stops the GET listening stream instead of reconnecting to
the same over-limit event.
## How Has This Been Tested?
New tests in `test/mcp/client/http_test.rb` cover the SSE and JSON rejection paths end-to-end,
the per-event reset (events that individually fit must not count against later ones),
the env-less `on_data` fallback, and the constructor validation. `bundle exec rake`
(tests, RuboCop, and conformance baseline) passes.
## Breaking Changes
Responses larger than 4 MiB are now rejected by default; pass a larger g`max_message_bytes:` to accept them.
## Motivation and Context
This repository does not track Gemfile.lock, so `bundle install` resolves to the latest RuboCop gems.
Upgrading rubocop-shopify from 2.x to 3.0.0 surfaces 8 new offenses because the shared config no longer
disables `Style/MutableConstant`, newly disables `Lint/RescueException` and `Naming/AccessorMethodName`,
and drops `AllowedPatterns` for `Naming/MethodName`. Upgrading rubocop itself to 1.88 surfaces one more offense
from the new `Style/ReduceToHash` cop.
This change fixes them as follows:
- Freeze mutable constants. Regexp constants are also flagged because `TargetRubyVersion` is 2.7,
where Regexp literals are not frozen.
- Remove inline disable directives that became redundant.
- Restore `AllowedPatterns: ['\Atest_']` for `Naming/MethodName` so `def test_` style test names
can reference class names such as `NoMethodError`.
- Swap out the frozen `SUPPORTED_STABLE_PROTOCOL_VERSIONS` constant in the protocol version fallback test
since Mocha cannot stub methods on frozen objects.
- Use `to_h` instead of `each_with_object` in `Server#index_resources_by_uri`.
Ref: https://rubygems.org/gems/rubocop-shopify/versions/3.0.0
## How Has This Been Tested?
`bundle exec rake` passes: 1402 runs, 0 failures, 0 errors, and RuboCop reports no offenses.
The conformance suite also passes against its baseline.
## Breaking Changes
None. Freezing these constants only affects code that mutates them in place, which has never been supported.
## Motivation and Context
The MCP Enterprise Managed Authorization extension (SEP-990, modelcontextprotocol/modelcontextprotocol#990,
accepted/final, published in the ext-auth repository) lets enterprise identity providers govern MCP authorization:
the client exchanges an IdP-issued ID token for an Identity Assertion Authorization Grant (ID-JAG) via
RFC 8693 token exchange at the IdP, then presents the ID-JAG to the MCP authorization server with
the RFC 7523 `jwt-bearer` grant, authenticating with `client_secret_basic`.
The `auth/cross-app-access-complete-flow` conformance scenario exercises the complete flow and previously failed
its two core checks (the Ruby client ran the authorization-code flow instead).
This mirrors the merged TypeScript SDK implementation (`requestJwtAuthorizationGrant` / `CrossAppAccessProvider` in
typescript-sdk; the Python SDK tracks the same work in the open python-sdk#1721), adapted to this SDK's provider design:
- New `MCP::Client::OAuth::IDJAGTokenExchange.request` performs the RFC 8693 exchange at the IdP token endpoint
(`subject_token_type` id_token, `requested_token_type` id-jag, `audience` = the MCP authorization server's issuer,
`resource` = the canonical MCP server URL) and validates that the response's `issued_token_type` is an ID-JAG
before returning the assertion. The ID-JAG itself is treated as opaque, as in the TypeScript SDK.
- New `MCP::Client::OAuth::CrossAppAccessProvider` declares `authorization_flow :jwt_bearer` and takes the assertion
from a callable (`call(audience:, resource:)`), so the common `IDJAGTokenExchange` case, enterprise secret stores,
and tests all plug in the same way (the TypeScript provider's assertion callback works identically).
The MCP AS credentials are stored as `client_secret_basic` client information.
- `Flow` dispatches `:jwt_bearer` alongside `:client_credentials`: discovery, issuer validation, and endpoint security checks
are shared with `run!`, then `run_jwt_bearer!` requests the assertion with the validated issuer as the audience and posts
the `jwt-bearer` grant through the existing `post_to_token_endpoint`.
- The conformance client builds a `CrossAppAccessProvider` when the harness injects `idp_id_token` (exchanging it at
`idp_token_endpoint` with `idp_client_id`, like the TypeScript and Python conformance clients),
and `auth/cross-app-access-complete-flow` is removed from `conformance/expected_failures.yml`.
## How Has This Been Tested?
- New `test/mcp/client/oauth/id_jag_token_exchange_test.rb` asserts the exact RFC 8693 form parameters on the wire
and the response validation: a non-ID-JAG `issued_token_type` is rejected, as are a missing `access_token`,
non-2xx responses, unparseable bodies, and non-object JSON.
- New `test/mcp/client/oauth/cross_app_access_provider_test.rb` covers the provider surface:
stored `client_secret_basic` credentials, the `:jwt_bearer` flow declaration, audience/resource passthrough to
the assertion callable, constructor validation (missing client_id, missing client_secret, non-callable assertion provider),
and storage delegation.
- New tests in `test/mcp/client/oauth/flow_test.rb` run the full grant against WebMock stubs: the token request carries
`grant_type=urn:ietf:params:oauth:grant-type:jwt-bearer`, the assertion, the RFC 8707 `resource`, and HTTP Basic credentials,
with no DCR and no authorization-code machinery; the assertion callable receives the validated issuer as `audience`
and the canonical server URL as `resource`; an empty assertion aborts before the token endpoint is contacted.
## Breaking Changes
None. `IDJAGTokenExchange` and `CrossAppAccessProvider` are new, and the `Flow` dispatch only activates for providers
that declare `authorization_flow :jwt_bearer`.
## Motivation and Context
A JSON-RPC message is an object, but two stdio-facing paths assumed that and raised
on a non-object value parsed from an incoming frame:
- `JsonRpcHandler.handle` maps each element of a batch array through `process_request`,
which reads `request[:id]` assuming a Hash. A non-object batch element (for example `[[]]`
or `[5]`) makes `[][:id]` raise `TypeError`. `handle_json` rescued only `JSON::ParserError`,
so the error escaped; over stdio it escaped the `StdioTransport#open` read loop.
The HTTP transport already rejects array bodies before dispatch, so this surfaced only on stdio.
- `MCP::Client::Stdio#read_response` parses each frame from the server and calls `parsed.key?("id")`,
which raises `NoMethodError` when the server emits a non-object frame.
Only `JSON::ParserError` was rescued, so one malformed line from the server escaped.
Both are robustness gaps: a malformed frame should be reported or skipped, not raise
an unhandled exception out of the transport.
- `process_request` now returns an `INVALID_REQUEST` error response for a non-object request
before touching `request[:id]`, so a malformed batch element becomes a normal per-element error.
- `handle_json` gains a last-resort `StandardError` rescue that returns an `INTERNAL_ERROR` response,
so an unexpected handling error still produces a JSON-RPC reply.
- `read_response` skips a non-object frame the same way it skips a frame without an id.
## How Has This Been Tested?
New tests in `test/json_rpc_handler_test.rb` cover non-object batch elements
(`[[]]`, `[5]`, `["x"]`, `[nil]`, `[true]`), a valid request followed by one non-object element,
and `handle_json("[[]]")` returning an error instead of raising. A test in
`test/mcp/server/transports/stdio_transport_test.rb` drives `open` with a `[[]]` line and
asserts it emits an error response without raising. A test in `test/mcp/client/stdio_test.rb`
has the server emit non-object frames before the response and asserts the client skips them
and still returns the response.
## Breaking Changes
None. A well-formed request, batch, or response is handled exactly as before;
only a non-object message now yields a JSON-RPC error on the server or is skipped by the client
instead of raising.
## Motivation and Context
Resolves the `elicitation-sep1034-client-defaults` client conformance scenario
(SEP-1034, https://github.com/modelcontextprotocol/modelcontextprotocol/issues/1034).
The Ruby HTTP client could not receive server-to-client requests at all:
it never opened the spec's standalone GET SSE listening stream, and SSE events that
were not responses were discarded. A server sending `elicitation/create` during a tools/call
(such as the conformance harness, which is built on the TypeScript SDK server)
either dropped the request silently or deadlocked the tool call waiting for an answer that
could never arrive.
This change adds three pieces:
### Server-to-client request dispatch
`MCP::Client::HTTP#on_server_request(method) { |params| ... }` registers a handler;
when a JSON-RPC request (a message with both `method` and `id`) arrives on any SSE stream,
the handler's return value is POSTed back to the server as the JSON-RPC `result`
(the server ACKs with 202 Accepted, per the transport spec).
Methods without a handler are answered with a `-32601` method-not-found error.
A handler may raise `MCP::Client::ServerRequestError` to answer with a specific JSON-RPC error code
(for example `-1` for a rejected sampling request, per the sampling spec),
mirroring the TypeScript SDK's `McpError` and the Python SDK's `ErrorData` return;
any other error is answered with a `-32603` internal error. In every case the server is not left waiting,
matching how the TypeScript and Python SDKs convert handler failures into JSON-RPC error responses.
### A standalone GET SSE listening stream
Registering a handler (once connected) opens the spec's listening stream on a background thread,
because servers deliver requests that are not tied to a client request on that stream -
the TypeScript SDK's `StreamableHTTPServerTransport` silently drops them when no GET stream is open.
Mirrors the TypeScript and Python SDK clients, which start listening right after the `initialize` handshake
without awaiting stream establishment. Reconnection semantics also follow both SDKs:
a gracefully closed stream is retried indefinitely after the server's `retry:` interval,
while consecutive connection failures (HTTP errors and idle read timeouts) stop the listener once
they reach the reconnection attempt cap, with a successful stream resetting the count;
a 405 stops immediately without retrying, since it is the spec's answer from a server that
offers no listening stream. The listening GET uses a 5-minute read timeout matching
the Python SDK's `sse_read_timeout` default. `close` terminates the thread.
### SEP-1034 defaults
`MCP::Client::Elicitation.apply_defaults` fills fields omitted from the elicitation content with
the `default` values declared in `requestedSchema` properties, never overwriting provided values
and skipping properties without a default. This follows the Python SDK, which applies defaults in
the elicitation callback; the TypeScript SDK's automatic `applyElicitationDefaults`
(gated on the `elicitation.form.applyDefaults` capability) is a possible follow-up.
`MCP::Client#on_elicitation` registers a handler for `elicitation/create` as a convenience.
The conformance client gains an `elicitation-sep1034-client-defaults` branch that connects with
`capabilities: { elicitation: {} }`, accepts the elicitation with defaults applied, and calls
the harness's `test_client_elicitation_defaults` tool; the scenario is removed from the expected failures baseline.
## How Has This Been Tested?
New unit tests covering: handler dispatch from a request's POST SSE stream with
the response body and 202 ACK verified, the `-32601` answer for unregistered methods,
the `-32603` answer for a raising handler, the custom-code answer when a handler raises
`ServerRequestError`, listener startup from both registration-after-connect and
connect-with-registered-handlers, listener shutdown after consecutive connection failures with
the count resetting after a successful stream and an immediate stop on 405,
and `apply_defaults` for all five SEP-1034 property types (string, integer, number, enum, boolean),
non-overwriting of provided values, omission of properties without defaults, and symbol-keyed schemas
## Breaking Changes
None. The listening stream and its background thread are created only when a server-request handler is registered;
clients that never call `on_server_request` / `on_elicitation` keep the existing single-request behavior.
## Motivation and Context
Both transports read a peer-controlled message fully into memory with no upper bound,
so a single unauthenticated message can allocate gigabytes and OOM the worker:
- `StreamableHTTPTransport#handle_post` did `request.body.read` with no size limit
before `JSON.parse`. A single ~512 MB POST grew a worker from 44 MB to 1.66 GB RSS
in the reporter's measurement. The path runs before session validation,
so it is reachable unauthenticated in both stateful and `stateless: true` modes.
- `StdioTransport` and `MCP::Client::Stdio` read frames with `IO#gets` and no `limit:`,
so a peer that never emits a newline grows a single String until the process is OOM-killed.
This matters when a host pipes an untrusted server's stdout into an unconstrained process.
The spec does not mandate a request size limit, and the reference SDKs vary
(the TypeScript SDK caps at a 4 MB default; Python/Go delegate to the web framework).
A finite default here bounds the allocation at the transport layer regardless of the deployment,
matching the TypeScript SDK's default.
- `StreamableHTTPTransport` gains `max_request_bytes:` (default 4 MiB). `handle_post`
rejects a declared `Content-Length` over the cap before reading, and also bounds
the actual read to one byte past the cap so a missing or spoofed `Content-Length`
(chunked transfer) is still caught, returning HTTP 413. `parse_request_body` now passes
`max_nesting:` (64) so a deeply nested body cannot exhaust the stack (complements the byte cap).
The constructor validates the new option as a positive Integer, since `nil` or a non-positive
value makes the underlying read unbounded and would silently disable the protection.
- The stdio `max_line_bytes:` bound (default 4 MiB) has since been merged upstream through
a separate advisory, so this change carries the remaining piece of the reported stdio fix:
an over-limit frame leaves leftover bytes in the pipe, desyncing the stream beyond recovery,
so `MCP::Client::Stdio` now closes the transport before raising and a later `send_request`
fails cleanly instead of parsing a truncated frame.
## How Has This Been Tested?
New tests in `test/mcp/server/transports/streamable_http_transport_test.rb` cover the 413 rejection
with and without `Content-Length`, the boundary body being accepted, the nesting-depth rejection,
and the constructor validation. The client stdio over-limit test now also asserts the transport
is closed afterward (`refute_predicate(transport, :connected?)`).
`bundle exec rake` (tests, RuboCop, and conformance baseline) passes. Verified end-to-end that
a 256 MB POST with no declared `Content-Length` is rejected with 413 without the transport reading
the whole body (a throttled `rack.input` that raises if read far past the cap is never tripped).
## Breaking Changes
None. The 4 MiB defaults comfortably fit a typical JSON-RPC message (a 4 MiB JSON string decodes to
roughly 3 MiB of base64 payload), and the new keyword arguments are additive and optional.
## Motivation and Context
CRuby's `IO#gets` without a limit reads until the next newline with no upper bound.
A peer that never emits a newline can therefore grow a single String until the process
is killed by the OS OOM killer.
This is a denial-of-service exposure (GHSA-7683-3w9x-ch42), identified during
a cross-SDK audit of the same unbounded-buffer pattern across the MCP SDKs.
It is most relevant when a host application spawns a sandboxed or untrusted MCP server
and pipes its stdout into an unconstrained host process, where a memory-capped peer
can exhaust the host.
Each frame is now capped at a configurable `max_line_bytes` (default 4 MiB,
large enough for any realistic frame including base64-embedded images,
and the value recommended in the advisory). A frame that reaches the limit without
a newline is treated as a transport error: the client raises, and the server reports
the error and closes the connection. A short final frame that ends at EOF without
a trailing newline is still accepted, so legitimate traffic is unaffected.
Both constructors validate `max_line_bytes:` as a positive Integer,
since `nil` or a non-positive value makes `IO#gets` read without an upper bound and
would silently disable the protection.
## How Has This Been Tested?
Server tests cover the `open` read loop stopping and reporting on an over-limit frame,
the `send_request` read loop raising on an over-limit response, and a valid final frame
without a trailing newline still being processed.
Client tests cover `read_response` raising on an over-limit response frame and
a normal response still round-tripping within a custom `max_line_bytes`.
Both transports also have tests rejecting invalid `max_line_bytes:` values
(`nil`, zero, negative, non-Integer) with an `ArgumentError`.
The full test suite and RuboCop pass.
## Breaking Changes
None. The default 4 MiB limit is far larger than any realistic JSON-RPC frame,
so existing well-behaved traffic is unaffected. The new `max_line_bytes:` keyword
on both transport constructors is purely additive and optional.
## Motivation and Context
SEP-2352 (modelcontextprotocol/modelcontextprotocol#2352, merged for the 2026-07-28 spec release) clarifies
authorization server binding and migration: clients MUST key persisted client credentials by the issuing
authorization server's issuer identifier, MUST NOT reuse credentials across different authorization servers,
and MUST re-register when the server's authorization server changes. CIMD `client_id`s (HTTPS URLs) are
explicitly portable across authorization servers, and a mismatch SHOULD surface an error rather than silently use
mismatched credentials.
The Ruby flow previously reused any stored `client_information` with a `client_id` unconditionally,
so a server that migrated to a new authorization server would have the old server's credentials
(and its dead tokens) replayed forever. This change follows the detect-change, invalidate,
and re-register approach that both reference SDKs converged on (typescript-sdk#2348/#2358, merged 2026-06-24;
python-sdk#2933/#2936/#2946, which uses the same proactive compare-before-reuse design as this change)
while keeping the 4-method storage contract intact by stamping the binding into the persisted hash:
- Successful Dynamic Client Registration now persists the credentials with an `"issuer"` member set to
the validated AS metadata `issuer` (which `ensure_issuer_matches!` has already pinned to the discovery URL).
- `Flow#ensure_client_registered` reuses stored credentials only when the stored issuer matches the current one.
Credentials without an issuer binding (data persisted by older SDK versions, or user-supplied pre-registered credentials)
are bound to the current issuer on first use. A portable CIMD `client_id` is reused and re-bound.
Any other issuer mismatch discards the stale registration and its tokens (tokens minted by the old AS are dead at the new one)
and re-registers via CIMD/DCR.
- `Flow#refresh!` raises `AuthorizationError` instead of replaying another authorization server's credentials at the token endpoint;
`MCP::Client::HTTP#attempt_refresh` already rescues that error and falls back to the full flow, which performs
the discard-and-re-register.
- The `client_credentials` grant (machine-to-machine) reads static pre-registered credentials without going through
`ensure_client_registered`, so it would have silently sent another authorization server's credentials after a migration.
`run_client_credentials!` now surfaces an `AuthorizationError` when the stored `issuer` mismatches the current authorization server,
implementing the spec's "SHOULD surface an error" (the TypeScript SDK raises `AuthorizationServerMismatchError` here).
Re-registration is not an option for pre-registered m2m credentials, so the operator must update the stored credentials;
credentials without a recorded issuer keep working unchanged.
Two adjacent hardenings from the reference implementations were checked and need no change here:
the legacy-fallback issuer stamp is self-consistent because this SDK's fallback registers at the same origin it
uses as the issuer (the cross-origin never-reached-AS binding python-sdk#2946 closed cannot arise),
and preserving a non-rotated `refresh_token` across refreshes already exists via `preserve_refresh_token`.
The TypeScript SDK's redirected-to-issuer callback gate is not adopted: the RFC 9207 `iss` validation (SEP-2468) covers that mix-up leg.
Resolves#385.
## How Has This Been Tested?
New tests in `test/mcp/client/oauth/flow_test.rb`:
- stored credentials with a matching issuer skip DCR
- an issuer mismatch re-registers and persists the new `client_id` with the new issuer
- the stale registration's tokens are cleared before re-registration (observed by aborting the flow at the DCR step)
- legacy credentials without an issuer (string- and symbol-keyed) are reused and bound to the current issuer on first use
- a CIMD `client_id` with a mismatched issuer is reused without DCR and re-bound
- fresh DCR persists the issuer binding
- `refresh!` succeeds with a matching stored issuer and raises (without contacting the token endpoint) with a mismatched one
- the `client_credentials` grant succeeds with a matching stored issuer and raises (without contacting the token endpoint) with
a mismatched one
All existing flow, HTTP OAuth, and provider tests pass unchanged.
## Breaking Changes
None for the documented storage contract: `client_information` hashes gain an additional `"issuer"` member that storages persist opaquely,
and credentials stored without one keep working via first-use binding. Behavior changes only in the previously broken case where
the authorization server changed, which now re-registers (or, for static machine-to-machine credentials, surfaces an error) instead of
replaying the old server's credentials.
## Motivation and Context
SEP-2549 (modelcontextprotocol/modelcontextprotocol#2549, merged for the 2026-07-28 spec release) adds a `CacheableResult`
shape with `ttlMs` (freshness lifetime in milliseconds, max-age semantics, 0 = do not cache) and `cacheScope`
(`"public"` or `"private"`, whether shared intermediaries may cache) to `ListToolsResult`, `ListPromptsResult`,
`ListResourcesResult`, `ListResourceTemplatesResult`, and `ReadResourceResult`, complementing the `listChanged` notifications.
The field shape follows the Python SDK's implementation (python-sdk#2824, `ttlMs` + `cacheScope` with `ttlMs: 0` /
`cacheScope: "public"` defaults; the TypeScript SDK's earlier `ttl`-seconds prototype #2070 was superseded by this shape).
Because this SDK negotiates protocol versions up to 2025-11-25, where the fields do not exist, emission is opt-in:
when neither `ttl_ms` nor `cache_scope` is configured, responses are serialized exactly as before. The fields become REQUIRED
in 2026-07-28; a follow-up can auto-emit defaults once that version enters `SUPPORTED_STABLE_PROTOCOL_VERSIONS`.
- `MCP::Server.new` gains `ttl_ms:` and `cache_scope:` keywords with validating writers modeled on `page_size=`
(`ttl_ms` must be nil or a non-negative Integer; `cache_scope` must be nil, `"public"`, or `"private"`).
- A private `apply_cache_metadata` wraps the five emission points. When the user sets only one of the two values,
the other is filled with the spec defaults (`ttlMs: 0`, the only universally safe freshness value, and `cacheScope: "public"`,
the spec and Python default).
- A `resources_read_handler` may return a full `{ contents:, ttlMs:, cacheScope: }` hash to override the server-level hints per result;
the documented bare-contents return shape keeps working unchanged.
- The client's `ListToolsResult` / `ListPromptsResult` / `ListResourcesResult` / `ListResourceTemplatesResult` structs gain
`ttl_ms` and `cache_scope` members populated from the response fields.
Resolves#387.
## How Has This Been Tested?
New tests in `test/mcp/server_test.rb`:
- with `ttl_ms: 5000`, all four list results carry `ttlMs: 5000` and the filled `cacheScope: "public"` default
- with only `cache_scope: "private"`, `resources/read` carries `ttlMs: 0`
- with neither configured, list and read results omit both fields (wire-format regression)
- the hints appear alongside `nextCursor` when paginating
- a `resources_read_handler` returning `{ contents:, ttlMs: 60_000 }` overrides the server-level value while the missing `cacheScope` is filled
- invalid writer values (`-1`, `1.5`, `"internal"`) raise `ArgumentError`
New tests in `test/mcp/client_test.rb` assert all four list result structs expose `ttl_ms` / `cache_scope`,
and that both are nil when the server omits the fields.
## Breaking Changes
None. Emission is opt-in, the new constructor keywords default to nil, and the client structs gain additive members.
## Motivation and Context
The MCP `io.modelcontextprotocol/oauth-client-credentials` extension
(SEP-1046, modelcontextprotocol/modelcontextprotocol#1046) specifies
two client authentication methods for the `client_credentials` grant:
`client_secret_basic` and the RECOMMENDED `private_key_jwt`
(RFC 7523 Section 2.2 JWT client assertion). The previous commit added
the grant with secret-based authentication; this adds the JWT method,
which is what the `auth/client-credentials-jwt` conformance scenario exercises.
This mirrors the merged implementations in the TypeScript SDK
(`PrivateKeyJwtProvider`, typescript-sdk#1157) and the Python SDK
(`PrivateKeyJWTOAuthProvider` / `SignedJWTParameters`, python-sdk#1663),
adapted to this SDK's provider design:
- New `MCP::Client::OAuth::JWTClientAssertion` builds the RFC 7523 compact
JWS with openssl, keeping the SDK free of a JWT gem dependency
(the TypeScript and Python SDKs use jose and PyJWT). Claims follow SEP-1046:
`iss` and `sub` carry the client_id, `aud` carries the authorization
server's issuer identifier, plus `exp` (300-second lifetime, matching both SDKs),
`iat`, and a unique `jti`. ES256 and RS256 are supported; ES256 signatures are
converted from OpenSSL's ASN.1 DER form to the raw 64-byte `r || s` encoding
JWS requires (RFC 7518 Section 3.4).
- `ClientCredentialsProvider` accepts
`token_endpoint_auth_method: "private_key_jwt"` with `private_key:`
(PEM string or `OpenSSL::PKey::PKey`) and a required `signing_algorithm:`
(no default, so an algorithm mismatch with the server's
`token_endpoint_auth_signing_alg_values_supported` fails loudly at construction;
both reference SDKs also take the algorithm explicitly).
The key stays on the provider and is never written to `storage`;
a throwaway assertion is signed at construction to fail fast on an unparseable key
or a key/algorithm mismatch.
- `Flow#post_to_token_endpoint` sends `client_assertion_type` and a freshly
signed `client_assertion` for `private_key_jwt`,
using the `ensure_issuer_matches!`-validated issuer as the audience,
and omits `client_id` from the body per SEP-1046
(identity travels in the assertion's claims).
- The conformance client builds a `private_key_jwt` provider
when the harness injects `private_key_pem`, defaulting the algorithm to
ES256 like the TypeScript and Python conformance clients,
and `auth/client-credentials-jwt` is removed from
`conformance/expected_failures.yml`.
Scope notes: only ES256 and RS256 are supported for now (the TypeScript and
Python SDKs accept whatever their JWT libraries support);
adding further algorithms is additive. There is also no injection point yet for
externally signed assertions (the TypeScript SDK's `StaticPrivateKeyJwtProvider` and
the Python SDK's `assertion_provider` callback cover KMS/HSM-held keys);
that is a separate follow-up.
## How Has This Been Tested?
New `test/mcp/client/oauth/jwt_client_assertion_test.rb` covers the JWS construction:
header and RFC 7523 claims, unpadded base64url segments,
ES256 signature round-trip verification (raw -> DER -> `OpenSSL::PKey::EC#verify`),
RS256 verification, PEM and parsed-key input, lifetime override, unique `jti` per assertion,
and rejection of unsupported algorithms, key/algorithm mismatches
(including a P-384 key with ES256), unparseable PEM, and public-only keys.
New tests in `test/mcp/client/oauth/client_credentials_provider_test.rb` cover
the provider surface: `private_key_jwt` construction, the key and
secret never reaching `client_information`, required `private_key` / `signing_algorithm`,
rejection of a simultaneous `client_secret`, fail-fast on key/algorithm mismatch,
and `client_assertion(audience:)` producing a JWT with the expected claims.
A new test in `test/mcp/client/oauth/flow_test.rb` runs the full grant against WebMock stubs
and asserts the token request carries `grant_type=client_credentials`,
the RFC 7523 `client_assertion_type`, an assertion whose `iss`/`sub` are the client_id
and whose `aud` is the issuer, the RFC 8707 `resource`, and neither `client_id` nor
any secret in the body or headers.
`bundle exec rake` passes; the `auth/client-credentials-jwt` conformance scenario
passes 7/7 checks and the baseline check passes with its expected-failure entry removed.
## Breaking Changes
None. `client_secret:` becomes optional on `ClientCredentialsProvider` only
in the signature; it is still required for the secret-based methods, and all
existing construction paths behave unchanged.
## Motivation and Context
SEP-2468 (modelcontextprotocol/modelcontextprotocol#2468, merged for the 2026-07-28 spec release) recommends
that authorization servers include the RFC 9207 `iss` parameter in authorization responses and requires MCP clients
to validate a present `iss` against the expected issuer before redeeming the authorization code.
This mitigates mix-up attacks in multi-IdP setups, where an attacker-controlled authorization server could otherwise capture
a code minted by an honest server.
This mirrors the TypeScript SDK's approach (typescript-sdk#2272, with typescript-sdk#1957 as the reference implementation),
adapted to the Ruby provider contract:
- `Provider`'s `callback_handler` may now return `[code, state, iss]` in addition to the existing `[code, state]`.
The 3-element form opts into validation; returning `iss = nil` asserts the caller inspected the authorization response
and found no `iss`.
- `Flow#run!` validates the issuer after the `state` check and before `exchange_authorization_code`,
implementing the RFC 9207 Section 2.4 rules: a present `iss` must equal the AS metadata `issuer` exactly
(the value `ensure_issuer_matches!` already pinned to the discovery URL); a nil `iss` from a 3-element callback is rejected
when the AS metadata advertises `authorization_response_iss_parameter_supported: true`; a legacy 2-element callback skips the check,
so every existing integration keeps working. On any violation the flow aborts with `AuthorizationError` before
the code reaches a token endpoint.
- The conformance client now returns the 3-element form, plumbing `iss` through from the redirect URL.
Resolves#386.
## How Has This Been Tested?
New tests in `test/mcp/client/oauth/flow_test.rb`:
- a matching `iss` completes the flow and persists tokens
- a mismatched `iss` raises `AuthorizationError` and never contacts the token endpoint (`assert_not_requested` on the token POST)
- a nil `iss` with `authorization_response_iss_parameter_supported: true` raises without contacting the token endpoint (fail closed)
- a legacy 2-element callback completes even when the AS advertises `iss` support (backward compatibility)
- a nil `iss` without the advertisement completes
- a mismatched `iss` is rejected even without the advertisement (validate-when-present is unconditional)
`bundle exec rake` (tests, RuboCop, and conformance baseline with the 3-element conformance callback) passes.
## Breaking Changes
None. The 2-element `callback_handler` contract continues to work unchanged; issuer validation activates only
when a callback opts in by returning the 3-element form.
## Motivation and Context
Resolves the `sse-retry` client conformance scenario (SEP-1699,
https://github.com/modelcontextprotocol/modelcontextprotocol/issues/1699).
Per SEP-1699, a server may close a request's SSE stream right after a priming event
(an event carrying an `id:`) without delivering the response, expecting the client
to treat the graceful close like a network failure: wait the server-specified
`retry:` interval, then reconnect with an HTTP GET carrying `Last-Event-ID`
so the server can replay the pending response on the resumed stream.
The Ruby client previously read the entire SSE body to EOF and parsed it afterward,
so it had no notion of per-event state (`id:`, `retry:`) or reconnection,
and the scenario failed all three checks (graceful reconnect, retry timing, `Last-Event-ID`).
`MCP::Client::HTTP#send_request` now consumes responses incrementally
via Faraday's `on_data` streaming callback, feeding SSE chunks to `event_stream_parser`
as they arrive while buffering plain JSON bodies as before. A new internal `SSEStream`
tracks the last received event id, the `retry:` reconnection delay,
and the awaited JSON-RPC response; once the response arrives, a held-open stream is
abandoned via an internal control-flow exception since servers may never close it.
When a stream closes gracefully after a priming event but before the response,
the client sleeps for the `retry:` interval (default 1000ms when the server sent none),
then issues a GET with `Accept: text/event-stream`, the session headers, and `Last-Event-ID`,
for up to 2 attempts. The defaults match the Python SDK's `DEFAULT_RECONNECTION_DELAY_MS` and
`MAX_RECONNECTION_ATTEMPTS` (`streamable_http.py`); the TypeScript SDK's `StreamableHTTPClientTransport`
(`streamableHttp.ts`) likewise lets the `retry:` field override its backoff and reconnects only when
a priming event was received and no response has arrived.
The public behavior of `send_request` is unchanged: same return values, same error mapping,
and a stream that closes without a priming event still raises the "No valid JSON-RPC response found in SSE stream" error
without reconnecting. Because the previous implementation read `response.body` and therefore worked with any Faraday adapter,
a fallback keeps that compatibility: when nothing was parsed during streaming, the body is read from `response.body`
(adapters without `on_data` support, e.g. the Faraday test adapter) or from the buffered chunks
(Faraday < 2.1, which invokes `on_data` without `env`).
The conformance client gains an `sse-retry` branch that calls the harness's `test_reconnection` tool,
and the scenario is removed from the expected failures baseline.
## How Has This Been Tested?
New unit tests covering: reconnection with `Last-Event-ID` after a primed graceful close including the `retry:` wait,
the default 1000ms delay when `retry:` is absent, raising after the reconnection attempt cap with `Last-Event-ID`
advancing across attempts, no reconnection for unprimed streams, and the non-streaming fallbacks
(JSON and SSE responses through the Faraday test adapter, which ignores `on_data`, plus buffered SSE chunks
when `on_data` receives no `env`)
## Breaking Changes
None. The `send_request` contract is unchanged; SSE bodies are now parsed incrementally instead of after EOF,
and reconnection only activates when the server opts in by sending a priming event.
## Motivation and Context
The MCP specification requires an `initialize` request followed by `notifications/initialized`
before any other interaction. Issue #334 makes that handshake explicit on the stdio client,
matching the Python SDK (`ClientSession.initialize()`) and TypeScript SDK (`Client.connect(transport)`).
The change was staged to avoid a hard break. Phase 1 (#336) added `MCP::Client::Stdio#connect`
while keeping an implicit-init compatibility shim on the first `send_request`.
Phase 2 (#338) emitted a deprecation warning when that shim ran. The warning has shipped since 0.16.0,
so this final phase removes the shim.
## How Has This Been Tested?
`send_request` now raises a `RuntimeError` when `connect` has not been called, and no longer
auto-starts the subprocess. `connect` is the sole entry point that spawns the process and
performs the handshake, which avoids leaving an orphaned subprocess behind.
The stdio test suite was reworked to call `connect` explicitly, a regression test was added for
the new guard, and `rake test` passes (1145 runs, 0 failures).
## Breaking Changes
Stdio clients that send requests without calling `MCP::Client#connect` now raise a `RuntimeError`
instead of silently initializing. Call `client.connect` before the first request.
Closes#334
## Motivation and Context
The MCP specification (SEP-2243) requires the Streamable HTTP transport to mirror
selected JSON-RPC body fields into HTTP headers so that intermediaries such as load balancers,
gateways, and observability tooling can route and inspect requests without parsing the body.
`Mcp-Method` carries the JSON-RPC `method` and is required on all requests and notifications,
while `Mcp-Name` carries `params.name` (`tools/call`, `prompts/get`) or `params.uri` (`resources/read`).
The Ruby SDK client did not send these headers, so servers enforcing SEP-2243 rejected
its requests with `400 Bad Request`, breaking interoperability. This aligns the Ruby SDK with
the TypeScript and Python SDKs, both of which send these headers from their Streamable HTTP clients.
Header values that cannot be transmitted safely as-is, such as non-ASCII text, control characters
(including CR/LF, which would otherwise allow header injection), or significant leading or
trailing whitespace, are wrapped per SEP-2243 as `=?base64?<base64>?=`. Safe ASCII values are sent unchanged.
The encoding uses `Array#pack("m0")` to avoid depending on the `base64` gem.
Resolves#366
## How Has This Been Tested?
Client tests cover deriving `Mcp-Name` from `params.name` (`tools/call`, `prompts/get`) and
falling back to `params.uri` (`resources/read`), both symbol- and string-keyed params,
requests and notifications without a name or uri sending only `Mcp-Method`, and non-Hash params sending
only `Mcp-Method`. Value encoding is covered for non-ASCII names, surrounding whitespace, CR/LF
(verifying no header injection), and values that already match the `=?base64?...?=` sentinel.
## Breaking Changes
None. The change is purely additive: existing requests now carry the required `Mcp-Method` and `Mcp-Name` headers
alongside the headers already sent. The `x-mcp-header` / `Mcp-Param-*` custom headers remain out of scope,
matching the current state of the TypeScript and Python SDKs.
## Motivation and Context
The server-side half of MCP `notifications/cancelled` shipped earlier and lets handlers
observe cancellation via `server_context.cancelled?` / `raise_if_cancelled!`.
This PR delivers the client-side equivalent that was deferred at the time:
an `MCP::Client` user can now cancel a request they have already issued and have
the calling thread wake up, matching the Python SDK
(`anyio.CancelScope`) and TypeScript SDK (`AbortSignal`) ergonomics.
The recommended pattern is to pass an `MCP::Cancellation` token into the request method,
run the request on a worker thread, and call `cancellation.cancel(reason:)` from another thread.
The cancelling thread sends `notifications/cancelled` to the server, and the calling thread is
woken up with `MCP::CancelledError`:
```ruby
cancellation = MCP::Cancellation.new
Thread.new do
client.call_tool(name: "slow_tool", arguments: {}, cancellation: cancellation)
rescue MCP::CancelledError
# cleanup
end
cancellation.cancel(reason: "user pressed cancel")
```
For low-level use, `Client#cancel(request_id:, reason:)` sends the notification without managing
the calling thread, and `Client#generate_request_id` lets callers pre-allocate an id before kicking off
a request on another thread. The `request_id:` and `cancellation:` keywords are accepted by
every request method (`tools`, `list_tools`, `resources`, `list_resources`, `resource_templates`,
`list_resource_templates`, `prompts`, `list_prompts`, `call_tool`, `read_resource`, `get_prompt`,
`complete`, `ping`).
When `cancellation:` is supplied, the actual blocking `transport.send_request` runs on a worker thread;
the calling thread waits on a `Queue` woken either by the response or by the cancel signal
(whichever arrives first - the same race contract as the server-side `StreamableHTTPTransport#cancel_pending_request`).
On a normal completion the `on_cancel` hook is deregistered so a late cancel does not emit a stray `notifications/cancelled`.
The worker thread is *not* force-killed when a cancel wins the race; it stays blocked on the underlying I/O until
the transport actually returns (or the user closes it). For `Client::HTTP` the leak resolves as soon as
the server sends any response; for `Client::Stdio` the user may need to call `client.transport.close`
if the server stops responding entirely.
`Client::Stdio` and `Client::HTTP` gain `send_notification(notification:)` so `Client#cancel` can deliver
the JSON-RPC notification through the existing transport plumbing. Custom transports that do not implement
`send_notification` cause `Client#cancel` to raise `NoMethodError`.
Ref: https://modelcontextprotocol.io/specification/2025-11-25/basic/utilities/cancellation
## How Has This Been Tested?
`test/mcp/client_test.rb` covers `Client#cancel` payload (with and without `reason`),
`Client#generate_request_id` returning distinct UUIDs, every request method threading `request_id:` through,
and the full `cancellation:` flow:
- Returns the response when no cancel happens (no behaviour regression for callers that do not pass a token).
- Pre-cancelled token raises `MCP::CancelledError` immediately and never reaches the transport
(no stray request, no stray notification).
- Mid-flight cancel raises `MCP::CancelledError` carrying the supplied reason and sends `notifications/cancelled`
through the transport.
- Late cancel after a normal completion does not emit a stray notification
(the `on_cancel` hook is deregistered in the `ensure` block).
`test/mcp/client/stdio_test.rb` covers `send_notification` writing the JSON line through the spawned process
and returning `nil` without waiting for a response.
`test/mcp/client/http_test.rb` covers `send_notification` POSTing the body, tolerating HTTP 202 Accepted,
and surfacing Faraday errors as `RequestHandlerError`.
## Breaking Change
None. All new keywords (`request_id:`, `cancellation:`) on request methods are optional and default to `nil`.
Existing callers that do not pass them get the original synchronous behaviour. Custom transports that already
implement `send_request(request:)` continue to work; only callers of `Client#cancel` need a transport
that responds to `send_notification`, and the failure mode is a clear `NoMethodError` rather than silent breakage.
## Motivation and Context
The MCP 2025-03-26 authorization spec predates Protected Resource Metadata:
the MCP server itself acted as the authorization base URL, its RFC 8414 metadata was fetched from
the server origin, and the spec's "Fallbacks for Servers without Metadata Discovery" section
required clients to use the default endpoints `/authorize`, `/token`, and `/register` relative to
the authorization base URL when no metadata was published. Both the TypeScript SDK
(`discoverOAuthServerInfo` falls back to the server base URL on PRM failure; `startAuthorization` /
`executeTokenRequest` / `registerClient` default the endpoint paths) and the Python SDK
(`build_oauth_authorization_server_metadata_discovery_urls(None, ...)` plus the same endpoint defaults)
keep this as client-side backwards compatibility, and the `auth/2025-03-26-oauth-metadata-backcompat` and
`auth/2025-03-26-oauth-endpoint-fallback` conformance scenarios exercise it.
The Ruby flow previously raised as soon as PRM discovery failed, so both scenarios were
expected conformance failures. This change adds the same fallback:
- `Flow#locate_authorization_server` first attempts PRM discovery; on any discovery failure
(404s, network errors, malformed documents, matching the TypeScript and Python SDKs' broad fallback)
the MCP server's origin becomes the legacy authorization base URL. The Communication Security check
(HTTPS or loopback) still applies to that origin, and PRM documents that parse correctly keep
their strict shape validation.
- `Flow#authorization_server_metadata` fetches RFC 8414 metadata from the base URL.
On the legacy path the `issuer` byte-match is skipped: the 2025-03-26 spec predates that requirement,
and a pre-PRM server may host its OAuth endpoints under a path prefix whose `issuer` legitimately differs
from the discovery origin (neither reference SDK validates the issuer on this path). On the modern path
the byte-match is unchanged.
- When even the metadata document is absent, synthetic metadata carrying the legacy spec's default endpoints
is used, with PKCE S256 assumed (the legacy spec mandates PKCE; the TypeScript and Python SDKs hardcode S256
on this path too). Endpoint HTTPS checks still apply.
- `run!` and `refresh!` share the new discovery helpers, and both 2025-03-26 scenarios are removed from
`conformance/expected_failures.yml`.
## How Has This Been Tested?
New tests in `test/mcp/client/oauth/flow_test.rb`:
- the metadata-backcompat shape: no PRM, AS metadata at the server origin with `/oauth`-prefixed endpoints
and a mismatched `issuer` completes the flow against those endpoints (no issuer error)
- the endpoint-fallback shape: no metadata at all registers, authorizes, and exchanges the code at `/register`,
`/authorize`, and `/token` on the origin, still sending `code_challenge` (S256) and `code_verifier`
- a remote plain-http origin is rejected as the legacy authorization base
- strict-mode regression: with PRM present, a mismatched `issuer` still aborts the flow
- a malformed PRM (top-level JSON array) now selects the legacy path instead of leaking a `TypeError`,
and surfaces a domain error when that path also dead-ends
Two existing tests that asserted PRM failure was terminal were updated to dead-end the legacy path explicitly,
preserving their original intent (single flow attempt, domain error type).
## Breaking Changes
None for spec-compliant 2025-06-18+ servers, whose PRM-based discovery and issuer validation are unchanged.
Servers that previously failed hard during PRM discovery now get one legacy discovery attempt before the flow errors,
which can only turn previously failing flows into working ones.
## Motivation and Context
SEP-837 (modelcontextprotocol/modelcontextprotocol#837, merged for the 2026-07-28 spec release)
updates the authorization spec to require MCP clients to specify an appropriate OIDC `application_type`
(`"native"` vs `"web"`) during Dynamic Client Registration, so the authorization server can apply
the matching redirect URI policy and native and web clients do not collide on redirect URI validation rules.
This mirrors the approach taken by the TypeScript SDK (typescript-sdk#2266) and the Python SDK (python-sdk#2784):
- `MCP::Client::OAuth::Discovery.infer_application_type(redirect_uris)` returns `"native"` when every redirect URI
is a native-app URI (a custom non-http(s) scheme per RFC 8252 Section 7.1, or an http(s) URI with a loopback host
per RFC 8252 Section 7.3) and `"web"` otherwise, including for nil, empty, or unparseable input.
Loopback detection reuses the existing `loopback_host?` helper, so lookalike hosts such as `localhost.example.com`
or `127.attacker.com` are not treated as native.
- `Flow#ensure_client_registered` now posts the metadata through the new `registration_client_metadata` helper,
which merges the inferred `application_type` into the DCR request body only when the user did not set
one explicitly (symbol or string key); an explicit value always wins.
Resolves#375.
## How Has This Been Tested?
New tests in `test/mcp/client/oauth/discovery_test.rb` cover the inference: loopback http URIs (IPv4, IPv6, `localhost`),
custom schemes, https URIs, mixed lists, the `localhost.example.com` lookalike, nil/empty input, and unparseable URIs.
New tests in `test/mcp/client/oauth/flow_test.rb` assert on the actual DCR request body via WebMock `assert_requested`:
loopback redirect URI registers `"application_type": "native"`, an HTTPS redirect URI registers `"web"`,
and an explicit `application_type` (symbol- or string-keyed) is never overridden.
## Breaking Changes
None. The DCR request body gains an `application_type` member only when the user has not specified one;
RFC 7591 requires authorization servers to accept and use known metadata fields, and OIDC-registration servers
already default this field to `"web"` when absent.
## Motivation and Context
The MCP authorization specification and SEP-1046 allow confidential clients to
obtain tokens through the OAuth 2.1 `client_credentials` grant: a machine-to-machine flow
with no user, no browser redirect, and no authorization code. The Ruby SDK
previously supported only the interactive authorization-code flow, so it could not
authenticate against a server that requires `client_credentials`. This makes
the `auth/client-credentials-basic` conformance scenario pass and is removed from
the expected failures list.
Following the TypeScript SDK (`ClientCredentialsProvider`) and the Python SDK
(`ClientCredentialsOAuthProvider`), the grant gets its own provider class rather than
overloading the authorization-code provider. The change adds:
- `MCP::Client::OAuth::ClientCredentialsProvider`, a dedicated provider
for the grant. `client_id` and `client_secret` are mandatory
(the grant is for confidential clients), `token_endpoint_auth_method` is
`client_secret_basic` (default) or `client_secret_post` -- `none` is
rejected because an unauthenticated `client_credentials` request is
meaningless. It has no redirect arguments, so a credentials-only
client never supplies the redirect URI / handlers that grant has no use for.
- `MCP::Client::OAuth::StorageBackedProvider`, an internal module that
shares the token / client-information persistence between the two provider classes
so `Flow` can treat any provider uniformly.
- `Provider#authorization_flow` (`:authorization_code`) and
`ClientCredentialsProvider#authorization_flow` (`:client_credentials`).
`Flow#run!` dispatches on this rather than inspecting
`client_metadata[:grant_types]`, which is protocol metadata for
the authorization server, not an SDK control signal.
- `run_client_credentials!` shares the same PRM / AS metadata discovery
and Communication Security checks as the interactive flow, then
exchanges `grant_type=client_credentials` at the token endpoint via
the existing `post_to_token_endpoint` (which already applies
`client_secret_basic`). There is no PKCE, redirect, or `offline_access`
augmentation, because the grant does not issue a refresh token
(OAuth 2.1 Section 4.3.3).
- `conformance/client.rb` builds a `ClientCredentialsProvider` for
the `auth/client-credentials-*` scenarios, reusing the harness-injected
`client_id` / `client_secret`.
## How Has This Been Tested?
New `ClientCredentialsProvider` tests cover: credentials are stored as
client information; `authorization_flow` is `:client_credentials`;
`client_secret_post` is accepted; missing `client_id` / `client_secret`
and `token_endpoint_auth_method: "none"` are rejected; and the token helpers
delegate to storage.
New `Flow` tests cover: a `client_credentials` provider exchanges
`grant_type=client_credentials` with HTTP Basic auth, saves the token,
and never contacts the authorization or registration endpoint;
the requested scope is taken from PRM `scopes_supported`; and a standard
`Provider` runs the authorization-code grant regardless of what its
`client_metadata[:grant_types]` lists. A `Provider` test asserts
`authorization_flow` is `:authorization_code`.
Conformance: `auth/client-credentials-basic` now passes 7/7 with no failures.
## Breaking Changes
None. `ClientCredentialsProvider` is a new class; the existing `Provider` keeps
its required `redirect_uri` / `redirect_handler` / `callback_handler` arguments
and its behaviour is unchanged. Existing interactive clients are unaffected.
## Motivation and Context
The MCP authorization specification (2025-11-25 scope-selection-strategy)
and RFC 6750 Section 3.1 describe OAuth 2.0 step-up authentication:
the client already holds a valid access token, but the server returns a 403
with `WWW-Authenticate: Bearer error="insufficient_scope", scope="..."`
to demand a higher-privileged scope for a specific operation.
The client is expected to run a fresh authorization request with
the escalated scope and retry.
Until now the Ruby SDK only ran the OAuth flow on 401 responses.
A 403 surfaced unchanged as a `RequestHandlerError` and the operation
that needed the escalated scope failed permanently, even with a valid token
in storage. This lined up with the `auth/scope-step-up` conformance
scenario being listed in `conformance/expected_failures.yml`.
The change adds:
- `MCP::Client::HTTP#send_request` recognises a 403 carrying
an `insufficient_scope` Bearer challenge as a step-up signal,
re-runs a full authorization request with the scope from the challenge,
and retries the original request. Retries are capped at
one per `send_request` call so a server that repeatedly demands more
scope cannot loop the transport.
- `run_step_up_flow!` deliberately bypasses the refresh-token path that
`run_oauth_flow!` uses on 401. Refreshing would re-issue the same
scope set the existing token already has and the server already
rejected. Only a fresh `authorization_code` grant with the escalated
scope can succeed.
- A plain 403 with no `insufficient_scope` challenge continues to
surface as `RequestHandlerError` with `error_type: :forbidden`.
- `conformance/client.rb` invokes a tool for the `auth/scope-step-up`
scenario; the scenario's authorization middleware only requires
the escalated scope on `tools/call`, so the previous `tools/list`-only
client could never trigger the second authorization.
- `auth/scope-step-up` is removed from `conformance/expected_failures.yml`.
Resolves#383.
## How Has This Been Tested?
New `HTTPOAuthTest` cases cover: the 403 insufficient_scope path
re-authorizes with the escalated scope from the challenge and retries
the original request with the new bearer token; a plain 403 without
the `insufficient_scope` challenge is surfaced as
`error_type: :forbidden` with no OAuth side effects; repeated
`insufficient_scope` responses retry at most once and then surface the 403;
a provider with a saved `refresh_token` still goes through
the full authorization request rather than exchanging the refresh token,
because the existing scope set is precisely what was rejected.
Conformance: `auth/scope-step-up` now passes 22/22 with no warnings.
`bundle exec rake test`, `bundle exec rake rubocop`, and
`bundle exec rake conformance` are all green.
## Breaking Changes
None. Behaviour on 401 is unchanged. A 403 without the `insufficient_scope`
Bearer challenge still raises the same `RequestHandlerError` as before.
The new retry only fires when both the WWW-Authenticate header signals step-up
and the transport was constructed with an OAuth provider.
## Motivation and Context
SEP-2351 (modelcontextprotocol/modelcontextprotocol#2351, merged for the 2026-07-28 spec release)
makes explicit that MCP uses the default `oauth-authorization-server` well-known URI suffix from
RFC 8414 Section 3.1 and defines no application-specific suffix.
`MCP::Client::OAuth::Discovery.authorization_server_metadata_urls` already conforms:
it probes the RFC 8414 default suffix first (path-inserted form for path issuers) and
otherwise uses only the OpenID Connect Discovery `openid-configuration` suffix.
Neither the Python SDK nor the TypeScript SDK needed a behavior change for this SEP either;
their tracking issues (python-sdk#2799, typescript-sdk#2256) remain documentation-level.
This change documents the SEP-2351 guarantee in the discovery helper and locks the behavior
in with regression tests so a future refactor cannot silently introduce a non-default suffix.
Resolves#384.
## How Has This Been Tested?
New regression tests in `test/mcp/client/oauth/discovery_test.rb`:
- the first probed candidate for a root issuer is exactly `/.well-known/oauth-authorization-server`
- every generated candidate uses only the `oauth-authorization-server` or `openid-configuration` suffix
- the RFC 8414 path-inserted `oauth-authorization-server` candidate comes first for issuers with a path component
- an issuer URL with a trailing slash is treated as a root issuer
## Breaking Changes
None. Documentation and regression tests only; no behavior change.
## Motivation and Context
SEP-2207 describes how an OAuth client obtains refresh tokens: it
declares the `refresh_token` grant type in its client metadata and
requests the `offline_access` scope, but only when the authorization
server advertises `offline_access` in its metadata `scopes_supported`.
Requesting `offline_access` when the server does not list it is forbidden.
The Ruby SDK already supports the refresh grant (`Flow#refresh!`) but never
requested `offline_access`, so it could not obtain a refresh token from a server
that gates it behind that scope. This aligns the SDK with the Python and TypeScript SDKs
and makes the `auth/offline-access-scope` conformance scenario pass
(previously it reported one warning, `sep-2207-client-metadata-grant-types`,
and was listed in the expected-failures baseline).
The change adds:
- `Flow#augment_scope_with_offline_access`, called from `run!` after
`resolve_scope`, appends `offline_access` to the requested scope when
both the client opted into refresh tokens (its registered
`grant_types` include `refresh_token`) and the authorization server
advertises `offline_access` in its metadata `scopes_supported`.
Already-present `offline_access` is not duplicated. Gating on the server
advertisement keeps the SDK from ever requesting the scope where it is not supported.
- `conformance/client.rb` declares `refresh_token` in `grant_types`, an honest reflection
of the SDK's refresh support, which clears the SEP-2207 grant-types warning.
- `auth/offline-access-scope` is removed from `conformance/expected_failures.yml`.
## How Has This Been Tested?
New `Flow` tests cover: `offline_access` is requested when the server advertises it
and the client declared the `refresh_token` grant; `offline_access` is NOT requested
when the client did not declare the grant; `offline_access` is NOT requested
when the server does not advertise it (even with the grant declared);
and `offline_access` is not duplicated when it is already part of the resolved scope.
Conformance: `auth/offline-access-scope` now passes 14/14 with no warnings,
and `auth/offline-access-not-supported` still passes 13/13 (no regression).
`bundle exec rake test`, `bundle exec rake rubocop`, and `bundle exec rake conformance`
are all green.
## Breaking Changes
None. `offline_access` is appended only when the client's own `grant_types` opt
into refresh tokens and the authorization server advertises the scope, so existing clients
that do neither see no change in the scope they request.
## Motivation and Context
The MCP 2025-11-25 authorization specification and `draft-ietf-oauth-client-id-metadata-document-00` define
Client ID Metadata Documents (CIMD): a client publishes its OAuth metadata at an HTTPS URL and uses that URL
as the OAuth `client_id` instead of going through Dynamic Client Registration. The Python and TypeScript SDKs
already implement this; the Ruby SDK now reaches parity.
The change adds:
- `MCP::Client::OAuth::Provider` accepts an optional `client_id_metadata_document_url:` keyword.
- A new `Discovery.client_id_metadata_document_url?` validator enforces structural requirements.
Spec-required: `https` scheme, a non-root path, and no fragment, userinfo, or `.`/`..` segments
(including the percent-encoded form `%2E`). Stricter than the draft as an SDK policy: query strings
are also rejected, because different encodings of the same query would yield distinct `client_id` strings for
the same logical document. The loopback `http` carve-out used for discovery URLs does not extend to CIMD URLs,
since the value is sent verbatim to the authorization server as the OAuth `client_id` and travels off-loopback.
Validation failures raise `Provider::InvalidClientIDMetadataDocumentURLError`.
- `Flow#ensure_client_registered` routes through CIMD only when the authorization server advertises
`client_id_metadata_document_supported` as JSON `boolean true`; truthy non-boolean values
(string `"true"`, `"false"`, etc.) fall back to DCR, matching the gating in the Python and TypeScript SDKs.
- The CIMD `client_id` is NOT persisted to storage. AS support is re-read on every flow so a server
that later drops CIMD support stops receiving the stale URL. Pre-registered `client_information`
continues to win over CIMD when both are available.
- `Flow#refresh!` reconstructs the CIMD `client_id` from live AS metadata when refreshing tokens obtained via CIMD.
When support has been withdrawn and no other identity is stored, it raises a clear `AuthorizationError` rather than
sending a `client_id` the authorization server no longer recognises.
- The Provider docstring and README OAuth section explain that DCR `client_metadata` and the JSON document served at
the CIMD URL are separate artifacts: DCR metadata MUST NOT include `client_id`, while the CIMD document MUST include
`client_id` set to the URL, `client_name`, and `redirect_uris` covering `redirect_uri`.
- `auth/basic-cimd` is removed from `conformance/expected_failures.yml`.
## How Has This Been Tested?
`Discovery` tests cover: https with path accepted; non-https rejected (http, ftp, file); root and empty path rejected;
fragment, query, and userinfo rejected; dot segments rejected including the percent-encoded form `%2E`; nil, empty,
and malformed inputs rejected.
`Provider` tests cover: the default `nil`; rejection of http URLs (loopback included); non-https schemes;
root and empty path; fragment; query; userinfo; and dot segments.
`Flow#run!` tests cover: skip DCR when the authorization server advertises CIMD and the provider has a URL;
fall back to DCR when the authorization server does not advertise support; fall back to DCR when the value is
the string `"false"` or `"true"`; pre-registered `client_information` wins over CIMD; the CIMD `client_id` is not
persisted to storage; a second flow against an authorization server that has dropped CIMD support recovers via DCR.
`Flow#refresh!` tests cover: refresh reconstructs the CIMD `client_id` without a stored `client_information`;
refresh raises when the authorization server no longer advertises CIMD support; stored `client_information`
takes precedence over the CIMD URL during refresh.
Conformance: `auth/basic-cimd` now passes 12/12 with no warnings. `bundle exec rake test`, `bundle exec rake rubocop`,
and `bundle exec rake conformance` are all green.
## Breaking Changes
None. The `client_id_metadata_document_url:` keyword is purely opt-in and defaults to `nil`.
When the keyword is omitted, the DCR flow is unchanged. Existing OAuth users see no behaviour change.
The CIMD path is only entered when both the user explicitly configures a URL and the authorization server
advertises support via `client_id_metadata_document_supported: true`.
## Motivation and Context
The MCP Authorization spec (2025-11-25) requires clients to support
Protected Resource Metadata discovery (RFC 9728), Authorization
Server Metadata discovery (RFC 8414), Dynamic Client Registration,
and OAuth 2.1 Authorization Code grant with PKCE (S256). The Ruby SDK
previously had no client-side OAuth support: the HTTP transport raised
`RequestHandlerError` on a 401 with no recovery, and 21 `auth/*` scenarios
were listed as known failures in `conformance/expected_failures.yml`.
This aligns the Ruby SDK with the Python and TypeScript SDKs,
both of which expose a pluggable OAuth client provider that drives
the full discovery + DCR + authorization + token exchange flow.
`MCP::Client::OAuth::Provider` mirrors that surface: callers supply
`client_metadata`, a `redirect_uri`, and pluggable `redirect_handler`
/ `callback_handler` hooks plus a token/client information storage.
`MCP::Client::OAuth::InMemoryStorage` is provided as a default.
`MCP::Client::HTTP` accepts a new optional `oauth:` keyword. When present,
the transport attaches `Authorization: Bearer` to every request,
and on a 401 parses `WWW-Authenticate` for the `resource_metadata` URL
and `scope` challenge, drives the discovery + DCR + authorization code flow
through the provider, and retries the failed request once with the acquired token.
Spec compliance details:
- The `WWW-Authenticate` parser handles multiple challenges
(e.g. `Basic ..., Bearer ...` or `Bearer ..., DPoP ...`), extracting only
the parameters that belong to the Bearer challenge.
- The PRM `resource` value is required to cover the MCP server URL
(same scheme/host/port, with the PRM path as a prefix of the server path)
before the flow continues, preventing redirection of issued tokens to a different audience.
- Both authorization and token requests always include a `resource`
parameter (RFC 8707), defaulting to a canonicalized form of
the MCP server URL when PRM does not advertise one.
- `client_secret_basic`, `client_secret_post`, and `none` token endpoint
authentication methods are supported. When pre-registered client information
specifies no `token_endpoint_auth_method`, the transport defaults to
`client_secret_basic` per RFC 6749 §2.3.1 an dthe TS/Python SDK convention.
Pre-registered client credentials supplied via `Storage#save_client_information`
with either string or symbol keys are honored, skipping DCR.
- Scope resolution follows the spec order: the `WWW-Authenticate` challenge scope wins first,
then PRM `scopes_supported`, and `Provider#scope` is the last-resort fallback only
if neither is supplied (the provider-supplied scope must not pre-empt server-advertised scope).
- Token endpoint error responses (RFC 6749 §5.2) are parsed: a 4x with
`error: "invalid_grant"` raises `Flow::InvalidGrantError`,
so `MCP::Client::HTTP` can discard the stored refresh token only when
the AS truly says it is dead. Generic `AuthorizationError`
(5xx, network failures, other error codes) leaves the refresh token
intact, so a transient AS outage does not force interactive reauth
on the next attempt.
- The OAuth `state` parameter is compared via
`OpenSSL.fixed_length_secure_compare` (with a `bytesize` guard) to
prevent timing-based discovery of the expected value.
- The 401 retry guard is a request-scoped local variable, so an MCP server
that keeps returning 401 is surfaced as an error after one
retry rather than triggering an infinite re-authorization loop.
## How Has This Been Tested?
Unit tests cover PKCE generation, multi-challenge `WWW-Authenticate` parsing,
PRM and Authorization Server metadata URL builders, the full Flow orchestrator
(success path, state mismatch, missing PKCE support, skipping DCR for both string
and symbol-keyed preregistered client information, `client_secret_basic`
token endpoint authentication, PRM `resource` mismatch rejection, fallback to
the server URL when PRM omits `resource`, scope resolution preferring PRM
`scopes_supported` over `Provider#scope`, default `client_secret_basic` for
confidential clients without an explicit `token_endpoint_auth_method`,
and the new `InvalidGrantError` vs generic `AuthorizationError` split for
token endpoint rejections), and the HTTP transport behavior
(401 recovery + retry, no-provider fallthrough, no infinite loop on repeated 401,
no infinite loop on flow failure, refresh-token retention on transient 5xx from
the token endpoint).
Conformance: `bundle exec rake conformance` newly passes 16 `auth/*`
scenarios beyond the existing baseline, including
`auth/metadata-default`, `metadata-var{1,2,3}`,
`scope-from-www-authenticate`, `scope-from-scopes-supported`,
`scope-omitted-when-undefined`, `scope-retry-limit`,
`token-endpoint-auth-{basic,post,none}`, `pre-registration`,
`resource-mismatch`, and `offline-access-not-supported`. Remaining
`auth/*` scenarios stay in `expected_failures.yml`.
`bundle exec rake test rubocop` is clean.
## Breaking Changes
None. `MCP::Client::HTTP#initialize` accepts a new optional `oauth:`
keyword that defaults to `nil`, preserving existing behavior. When unset,
the transport raises `MCP::Client::RequestHandlerError` on a 401 exactly as before.
## Motivation and Context
This commit implements Phase 1 of 3 for #334.
The MCP specification requires an `initialize` request followed by
a `notifications/initialized` notification before any other interaction.
PR #327 added `MCP::Client#connect` and `MCP::Client::HTTP#connect`
to expose this handshake explicitly for the HTTP transport,
but stdio remained on its private `initialize_session` lazy-init path,
inconsistent with the Python SDK (`ClientSession.initialize()`) and
TypeScript SDK (`Client.connect(transport)`),
both of which require an explicit handshake call regardless of transport.
The implicit init path on the first `send_request` is
preserved as a non-breaking compatibility shim. Public API added on `Stdio`:
- `connect(client_info:, protocol_version:, capabilities:)` performs the
handshake, validates the negotiated `protocolVersion` against
`MCP::Configuration::SUPPORTED_STABLE_PROTOCOL_VERSIONS`,
caches the `InitializeResult`, and is idempotent.
- `connected?` reports handshake completion.
- `server_info` exposes the cached `InitializeResult`.
Documentation and examples are updated so new users learn the explicit pattern
from the start.
## How Has This Been Tested?
Fifteen regression tests cover: explicit handshake, `server_info` caching,
idempotence, custom keyword arguments, JSON-RPC error response, missing
or non-Hash `result`, unsupported protocol version with state rollback,
state rollback when the `notifications/initialized` write fails,
retry after a failed handshake, and the `connected?` / `server_info` lifecycle.
The pre-existing stdio tests pass without modification, verifying that
the implicit-init compatibility path still works.
## Breaking Changes
None for the typical "implicit init via `client.tools` / `client.call_tool`" usage.
Existing stdio clients keep working unchanged.
Three observable behavior changes for code that inspects transport state directly:
- `Client#connected?` for a never-initialized stdio client returns `false`
(previously fell through to a `true` default via the `respond_to?` fallback).
- `Client#server_info` for stdio returns the cached `InitializeResult`
after the handshake (previously always `nil` via the fallback).
- The implicit handshake path now validates the server's negotiated
protocol version, raising `RequestHandlerError` for unsupported versions.
Production servers that return one of the supported versions are unaffected;
this closes a pre-existing spec-conformance gap on stdio.
Per MCP spec, clients that no longer need a session SHOULD send an HTTP
DELETE to the MCP endpoint with the Mcp-Session-Id header to explicitly
terminate it. Servers MAY respond with 405 Method Not Allowed when they
don't support client-initiated termination.
`MCP::Client::HTTP#close` sends the DELETE with session headers, clears
local session state regardless of outcome, and is a no-op when no
session has been established.
https://modelcontextprotocol.io/specification/2025-11-25/basic/transports#session-management
Per the Streamable HTTP spec, if the server returns an Mcp-Session-Id
header during `initialize`, the client MUST include it on subsequent
requests. Capture the session ID and protocol version from the
`initialize` response and attach them automatically on subsequent
POSTs.
Per the Protocol Version Header section of the spec, the client MUST
also include `MCP-Protocol-Version` on all subsequent requests so the
server can respond based on the negotiated protocol version.
Map 404 responses to a new `SessionExpiredError` (a subclass of
`RequestHandlerError` for backward compatibility) and clear local
session state so callers can start a fresh session with a new
`initialize` request, as required by the spec.
- https://modelcontextprotocol.io/specification/2025-11-25/basic/transports#session-management
- https://modelcontextprotocol.io/specification/2025-11-25/basic/transports#protocol-version-header
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Per the Streamable HTTP spec, a server returns 202 Accepted with no
body as the ACK for a JSON-RPC notification or response. The client
previously errored on 202 because it fell through Content-Type
dispatch. Short-circuit to return nil before Content-Type handling so
fire-and-forget messages (e.g. notifications/initialized) no longer
raise.
https://modelcontextprotocol.io/specification/2025-11-25/basic/transports#sending-messages-to-the-server
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The Streamable HTTP spec allows servers to respond with either
`application/json` or `text/event-stream`. The client previously
rejected the latter. Add SSE parsing via the optional
`event_stream_parser` gem (users add it to their Gemfile when their
server uses SSE), dispatch on the response Content-Type, and scan the
event stream for the first JSON-RPC response message.
Document the new optional dependency in both the README's HTTP
Transport Layer section and docs/building-clients.md.
No other Streamable HTTP features (202 handling, session IDs,
`connect`, DELETE termination) are included here; those will land in
follow-up PRs.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
## Motivation and Context
The MCP specification defines cursor-based pagination for list operations that
may return large result sets:
https://modelcontextprotocol.io/specification/2025-11-25/server/utilities/pagination
Pagination allows servers to yield results in smaller chunks rather than all at once,
which is especially important when connecting to external services over the internet.
The Ruby SDK previously returned complete arrays for all list endpoints
(`tools/list`, `resources/list`, `prompts/list`, `resources/templates/list`)
without pagination support. This adds cursor-based pagination.
### Server-side
A new `MCP::Server::Pagination` module is introduced (mixed into `MCP::Server`)
that provides the `paginate` helper for slicing a collection by cursor and
a `cursor_from` helper that extracts the cursor from the request params
while rejecting non-Hash inputs with `-32602 Invalid params` per the specification.
`MCP::Server.new` accepts a new `page_size:` keyword argument.
When `nil` (the default), all items are returned in a single response,
preserving the existing behavior. When set to a positive integer,
list responses include a `nextCursor` field if more pages are available.
`Server#page_size=` is a validating setter that rejects anything other than
`nil` or a positive `Integer`, raising `ArgumentError` on invalid inputs.
The constructor routes through the setter, so both `Server.new(page_size: 0)`
and `server.page_size = -1` raise.
Cursors are string tokens carrying a zero-based offset, treated as opaque by
clients. Cursor inputs are validated to be strings per the MCP spec;
invalid cursors (non-string, non-numeric, negative, or out-of-range) raise
`RequestHandlerError` with error code `-32602 (Invalid params)` per the spec.
### Client-side: two complementary APIs
The SDK exposes two API shapes for listing, each serving a different use case:
1. **Single-page with cursor** (`client.list_tools(cursor:)`,
`list_resources`, `list_resource_templates`, `list_prompts`): each call issues
one JSON-RPC request and returns a result object (`MCP::Client::ListToolsResult` etc.)
carrying the page items and an optional `next_cursor`.
These methods match the single-page-with-cursor convention used by the Python SDK
(`session.list_tools(params=PaginatedRequestParams(cursor=...))`)
and the TypeScript SDK (`client.listTools({ cursor })`). The method name and
the `cursor` parameter name are identical across the three SDKs,
so pagination code translates directly between languages.
2. **Whole-collection** (`client.tools`, `client.resources`, `client.resource_templates`,
`client.prompts`): auto-iterates through all pages and returns a plain array of items,
guaranteeing the full collection regardless of the server's `page_size` setting.
The auto-pagination loop tracks previously seen cursors in a set and exits if the server
revisits any of them, guarding against both immediate repeats and multi-cursor cycles
(e.g. `A -> B -> A`). This mirrors Rust SDK's `list_all_*` convenience methods,
though the Ruby names are preserved from the pre-pagination 0.13.0 API for backward compatibility.
Future rename to `list_all_*` is left as a TODO in the source.
Result classes are `Struct`s named to mirror Python SDK's `ListToolsResult` /
`ListPromptsResult` / etc. to align naming with other SDKs. Each struct exposes
its page items (e.g. `result.tools`), an optional `next_cursor` for continuation,
and an optional `meta` field mirroring the MCP `Result` type's `_meta` response field
so servers that decorate list responses do not lose metadata through the client.
### Ruby-specific ergonomics
`Server.new(page_size:)` and the `Pagination` module are Ruby-specific.
The Python and TypeScript SDKs do not expose a built-in `page_size` helper;
developers there re-implement cursor slicing inside each handler.
In Ruby, setting `page_size:` on the server is enough for the SDK to handle
the slicing across the four built-in list endpoints.
### README
A Pagination section documents both the server-side `page_size:` option and
the client-side iteration patterns, including an explicit note that each `list_*` call
is a single JSON-RPC round trip whose response size depends on the server's `page_size`,
and that `client.tools` / `.resources` / `.resource_templates` / `.prompts` return
the complete collection when needed.
## How Has This Been Tested?
- Added tests for the `Pagination` module (`test/mcp/server/pagination_test.rb`),
server-side pagination (`test/mcp/server_test.rb`), and client-side pagination
(`test/mcp/client_test.rb`).
- Added and existing tests pass. `rake rubocop` is clean and `rake conformance` passes.
## Breaking Changes
None. This is a new feature addition. Existing code continues to work unchanged:
`page_size` defaults to `nil` on the server (returns all items, no `nextCursor`),
and the client's existing `tools` / `resources` / `resource_templates` / `prompts` methods
continue to return complete arrays by transparently iterating through pages.
## Motivation and Context
`MCP::Client::HTTP` builds its Faraday connection internally in a private `client` method
with no way to customize the middleware stack or adapter.
Users who need observability (request/response recording, failure logging) or
a different HTTP adapter must override a private method, coupling to an internal API.
This accepts an optional block in `MCP::Client::HTTP.new` that yields the Faraday builder
after default middleware is configured, allowing users to add custom middleware or swap
the HTTP adapter.
## How Has This Been Tested?
Added a test that verifies custom headers set via the block are included in requests.
All existing tests continue to pass.
## Breaking Changes
None. The block argument is optional, so existing code is unaffected.
Resolves#303
## Motivation and Context
The `MCP::Client` currently only supports HTTP transport (`MCP::Client::HTTP`).
To achieve parity with other MCP SDKs (Python, TypeScript, C#) which provide
stdio client transports for spawning and communicating with server subprocesses,
the Ruby SDK needs a stdio client transport.
https://modelcontextprotocol.io/specification/2025-11-25/basic/transports#stdio
This enables building MCP clients that launch server processes and communicate
via stdin/stdout using newline-delimited JSON-RPC 2.0 messages.
## How Has This Been Tested?
Added tests for process spawning, protocol initialization handshake,
notification handling, error handling, and transport close lifecycle.
## Additional context
Enabled the `Lint/IncompatibleIoSelectWithFiberScheduler` cop in `.rubocop.yml`
to detect accidental use of `IO.select` instead of `IO#wait_readable`.
