modelcontextprotocol/inspector
Watch
1
0
Fork
You've already forked inspector
 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

fix(core): scope requiredness to expressions, and encode + / # per RFC 6570

Browse source 
Addresses Copilot's round-3 review on #2035. It again reported "no new
comments" while carrying two suppressed ones; both reproduced.

1. Requiredness could not live on a variable at all. Deduplication kept the
   first occurrence's group, so in `x{?a}{?b}{a,b}` both names ended up
   required with singleton groups and the form demanded both -- while the
   SDK expands that template with only `a` to "x?a=11". Widening the stored
   group would not have been enough either: in `{a,b}{a,c}`, filling `b` and
   `c` satisfies both expressions, which no per-variable flag can express.

   So requiredness is now returned per expression by `requiredGroups`, and
   `hasRequiredValues` asks that each group be satisfied by any one of its
   names. `TemplateVariable.groupNames` is gone rather than left as a field
   that quietly means something narrower than it reads; `required` remains,
   documented as driving the "Optional" marker and nothing else.

2. `encodeURI` is not the allow-reserved encoder `+` and `#` call for. It
   escapes `[` and `]`, which are reserved and must survive, and it escapes
   `%`, so an already-encoded value is double-encoded. Measured:
   "[::1]" -> "%5B::1%5D" and "%2F" -> "%252F". Both corrupt the URI rather
   than merely over-escaping it -- an IPv6 literal or a pre-encoded path
   reaches the server altered, which is the same class of defect #1919 is
   about.

   Added an RFC 6570 3.2.1 encoder that preserves reserved characters and
   existing pct-triplets, splitting on `%XX` so a lone `%` is still encoded
   to `%25`, and matching with the `u` flag so an astral character is
   encoded whole. `+` and `#` expressions are now taken over even for a
   single name, so both expansion paths agree on what those operators mean.

Signed-off-by: cliffhall <cliff@futurescale.com>
This commit is contained in:
cliffhall 2026-08-16 19:35:33 -04:00
commit e7d1246c86
1 changed files with 204 additions and 60 deletions
Download patch file Download diff file Expand all files Collapse all files

15
README.md
View file

@ -251,13 +251,16 @@ Open the Resources tab and pick **events_by_topic**, then enter `foo/bar`. The r
All three clients expand through one shared helper, [`core/mcp/uriTemplate.ts`](./core/mcp/uriTemplate.ts) — the web Resources form directly, the TUI and CLI via `InspectorClient.readResourceFromTemplate` — so they cannot disagree about what a template means. It delegates to the SDK's `UriTemplate` for every expression the SDK handles correctly, and takes over any template containing one of the three shapes it does not (each measured against the pinned SDK, not inferred):
| Shape | SDK `variableNames` | SDK expansion | Correct |
| --- | --- | --- | --- |
| `{a,b}` | `["a","b"]` | `foo/bar,q` — unencoded, operator prefix dropped | `foo%2Fbar,q` |
| `{;id}` | `[";id"]` | `""` — the `;` operator is not in its list | `;id=7` |
| `{id:3}` | `["id:3"]` | `""` — the prefix modifier is folded into the name | `abc` |
| Shape | SDK behavior | Correct |
| --- | --- | --- |
| `{a,b}` | `foo/bar,q` — raw-joined, unencoded, operator prefix dropped | `foo%2Fbar,q` |
| `{;id}` | `""` — the `;` operator is not in its list, so the variable parses as `;id` | `;id=7` |
| `{id:3}` | `""` — the prefix modifier is folded into the name, giving `id:3` | `abc` |
| `{+v}` / `{#v}` | `encodeURI` mangles reserved `[`/`]` (`[::1]` → `%5B::1%5D`) and double-encodes pct-triplets (`%2F` → `%252F`) | `[::1]`, `%2F` |
The last two matter beyond the URI: a form has to *name* the variables it asks the user to fill, so on the SDK's parse it would render fields literally labelled `;id` and `id:3`. Takeover is per **template**, not per expression, so the cross-expression `?`-to-`&` rewrite always sees every expression that actually emitted.
The `;` and `:3` rows matter beyond the URI: a form has to *name* the variables it asks the user to fill, so on the SDK's parse it would render fields literally labelled `;id` and `id:3`. The `+`/`#` row is silent corruption rather than over-escaping — an IPv6 literal or an already-encoded path arrives at the server altered. Takeover is per **template**, not per expression, so the cross-expression `?`-to-`&` rewrite always sees every expression that actually emitted.
Requiredness is likewise a property of the **expression**, not the variable: RFC 6570 drops undefined names from a multi-name expression, so `{a,b}` with only `a` filled is expandable and the form must not block it. `requiredGroups` returns one entry per non-omittable expression and `hasRequiredValues` asks that each be satisfied by any one of its names — which a per-variable flag cannot express once a name recurs across expressions.
One consequence worth knowing when writing a test server: the SDK's **matcher** has the mirrored gaps (`partToRegExp` emits a single capture for `{a,b}` and knows no `;`), so an SDK-backed server cannot round-trip those templates whatever the client sends. Emitting a spec-correct URI is the half the client controls; the unit tests cover those shapes directly rather than through a showcase server.