* chore: land the #2247 round-9 review findings on v2/main Copilot filed a ninth review round on #2247 after it had merged, so none of it reached v2/main. All five findings are addressed here. 1. The testing -> test-servers hand-off delivered the wrong half of the procedure. test-servers documented only the two-process manual path (server-composable.js --config <name>.json plus a second Inspector process), but an integration test builds the server in-process from the @modelcontextprotocol/inspector-test-server API and owns its lifecycle. Adds a "Two ways to use a fixture — pick the right one first" section with the in-process recipe, the four mechanics that bite (one barrel, start/stop in afterEach, read started.url, era is a constructor option), and a pointer to the reference test; renames "Run one" to "Run one by hand (two processes)" so the manual path names itself. testing now says in the connecting bullet that an automated test does NOT spawn. 2. testing's heading asked "does the test need a real server?", which its own hand-rolled-server exception contradicts. It now asks whether the test uses a test-servers/ fixture. 3. "imports" was too broad a dependency test — test-server-scope.test.ts imports the package to unit-test scope middleware as a pure function and never starts a server. The condition is now "starts, spawns, or configures", with that test named as the counterexample. 4. docs/skill-authoring.md attributed the 33% -> 100% rise to imperative wording and placement independently. The #2247 edit changed both at once, so only the combination is established; the split into two levers is now labelled the plausible mechanism, not a measured result. 5. The CHAIN_THRESHOLD comment generalised "a well-shaped pointer clears 0.8" from one two-case run of that same multi-variable edit. It now reads as an existence proof about the ceiling, not a property of pointers in general. Also records the headline lesson where it can be found again: a green hand-off number proves skill B loaded, never that B answered the question — in AGENTS.md's chain-case rules and in the skill-authoring checklist. Docs, skill bodies and one comment only; no runtime code changes. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_018zwYWanPhvgu6wiKoeGg4N Signed-off-by: cliffhall <cliff@futurescale.com> * fix: address Copilot review round 1 on #2264 The new section claimed two shapes and there are three — the automated stdio path was missing, and it is the repo's established pattern (getTestMcpServerCommand() into a stdio transport, used by five web integration files and the whole CLI suite). Telling callers an automated test "does not spawn" told them not to use it. - test-servers: "Two ways" -> "Three ways", with a table and a subsection for the spawned-stdio shape — the transport owns the child (disconnect, not stop()), no config can be selected because that entry point runs the stdio default, and it is still the build. Names a reference test for it. - test-servers: the build is no longer scoped to the manual path. All three shapes resolve test-servers/build/ — two through the barrel alias, one by running the emitted .js — so the staleness hazard is called out as applying to every shape, and `Build first` no longer opens by claiming the only reason is subprocess spawning. - testing: the "does not spawn" instruction is replaced by the two automated shapes and when to pick each. What they share, and what the hand-off actually needed to say, is that neither picks a showcase config. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_018zwYWanPhvgu6wiKoeGg4N Signed-off-by: cliffhall <cliff@futurescale.com> * fix: address Copilot review round 2 on #2264 Round 1 fixed the missing stdio shape but kept the wrong axis. Cutting the taxonomy by "automated vs by hand" excluded the config-driven smokes — smoke-web-elicitation.mjs spawns server-composable.js --config directly, and lib/mcp-app-flow.mjs does it for smoke:web:app, smoke:web:tabs and pack:verify. Those are automated AND need every bit of the config and protocol-era guidance a person at two terminals does, so the skill was routing their authors away from it. - test-servers: the cut is now how the SERVER IS STOOD UP, stated as such. Third shape is "spawned composable HTTP", whose consumers are the smokes, pack:verify, and a person by hand — the config/era guidance belongs to the shape, whoever starts it. Table gains a Consumers column and loses the Who column that caused this. - test-servers: the stale "an automated test builds the server in-process instead" line under `Run one by hand` is gone (suppressed comment); that heading is now the same shape driven by a person. - testing: routes by what is driving — integration/CLI test to in-process HTTP, stdio-is-the-point to spawned stdio, and a smoke to the composable-config path with the era guidance it actually needs. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_018zwYWanPhvgu6wiKoeGg4N Signed-off-by: cliffhall <cliff@futurescale.com> * fix: address Copilot review round 3 on #2264 Round 2 replaced one wrong axis with a narrower one: it routed "a smoke" wholesale to the config-driven composable path, and smoke:cli does not use it. smoke-cli.mjs starts an in-process createTestServerHttp for the header round-trip (so it can read back what the CLI sent) and writes the built stdio entry into a --catalog for the connect checks; smoke:tui uses only the latter. Neither picks a config. - test-servers: consumers filled in per row — in-process HTTP gains CLI tests and smoke:cli, spawned stdio gains smoke:cli and smoke:tui, and the composable row is narrowed to the WEB smokes. A warning states the point directly: "a smoke" is not a shape, pick by the row. - test-servers: the stdio shape now covers naming the built entry in a catalog without connecting, which is what those two smokes do. - testing: same narrowing, plus the reason spelled out so the word "smoke" stops being a routing key. - testing: the dependency criterion said "starts, spawns, or configures", which excluded the build-only smoke:tui case documented directly above it — the skill both required and rejected the hand-off for it (suppressed comment). It now reads "starts, spawns, configures, or hands a built entry to the subject under test". Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_018zwYWanPhvgu6wiKoeGg4N Signed-off-by: cliffhall <cliff@futurescale.com> --------- Signed-off-by: cliffhall <cliff@futurescale.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
Cliff Hall
GitHub
parent
d8b4b06af9
commit
ab80221ce6
1 changed files with 250 additions and 18 deletions
|
|
@ -201,6 +201,17 @@ ever taken.** `test-servers` scores 5/5 on its own cases and every one of them
|
|||
asks for it by name; a skill only ever reached _through_ another would score a
|
||||
clean 100% while the hand-off silently never fired (#2204).
|
||||
|
||||
⚠️ **A green hand-off number is not evidence the hand-off is USEFUL.** The case
|
||||
can observe one thing — that skill B was loaded. Whether B then answers the
|
||||
question the prompt actually asked is outside what any `chain` case measures, in
|
||||
either direction: B can be loaded and say nothing relevant, or say something
|
||||
actively wrong for the caller that arrived through A. `testing` ->
|
||||
`test-servers` scored 100% while `test-servers` documented only the two-process
|
||||
manual path, so every model that followed the pointer from an *integration-test*
|
||||
prompt was handed the wrong half of the procedure (#2264). So when a chain case
|
||||
goes green, **read B's body as the caller who arrives through A** and check that
|
||||
what it says fits that caller. The number cannot do that for you.
|
||||
|
||||
A chained case names the ordered skills one run should load:
|
||||
|
||||
```json
|
||||
|
|
@ -285,8 +296,8 @@ license (Copilot). A strict bound of `1.0` is therefore unreachable and the
|
|||
harness rejects it up front rather than failing every case.
|
||||
|
||||
0.5 is a floor on **useful reliability for a second-hop load**, not a claim that
|
||||
0.8 is out of reach — a well-shaped pointer clears it outright, at 100% in the
|
||||
worked example below. Note that a *chain* bar of 0.8 would be a **strict** one
|
||||
0.8 is out of reach — the one pointer measured after reshaping cleared it
|
||||
outright, at 100% in the worked example below. Note that a *chain* bar of 0.8 would be a **strict** one
|
||||
(`> 0.8`, the comparison this threshold uses; the first-move 0.8 is the
|
||||
inclusive `>=`), so at `RUNS=5` only a clean 5/5 would pass it — 4/5 would not. What 0.5 buys is that the column keeps carrying signal across the
|
||||
*range* of pointer strengths a repo actually has: a hand-off is a noisier
|
||||
|
|
@ -319,11 +330,18 @@ measured 100% / 80% on the full suite and 100% / 100% on a focused
|
|||
noise, well short of the 67 above.)
|
||||
|
||||
The transferable part is that **a pointer is followed when it reads as an action
|
||||
with a trigger, and skimmed when it reads as a fact.** #2202 found the same
|
||||
lever on a *description*'s shape; this is it applied to a body. The corollary is
|
||||
where to put one: the top of a body is read before the model knows it needs the
|
||||
second skill, so a pointer that lives only there is a pointer it has already
|
||||
scrolled past by the time it matters.
|
||||
with a trigger, placed where the decision is actually made.** #2202 found the
|
||||
same lever on a *description*'s shape; this is it applied to a body.
|
||||
|
||||
⚠️ **Read that as ONE lever, not two.** The #2247 edit changed the wording *and*
|
||||
the placement in the same revision, and only the combination was measured, so
|
||||
the run establishes the pair — it does not license attributing the rise to
|
||||
imperative phrasing alone, nor to repetition alone. The split into two levers is
|
||||
the plausible reading of the mechanism, not a measured result: the top of a body
|
||||
is read before the model knows it needs the second skill, so a pointer that
|
||||
lives only there has plausibly been scrolled past by the time it matters. Taking
|
||||
them apart would need two more runs — reword without moving, and move without
|
||||
rewording — and nobody has paid for those. Until someone does, ship both.
|
||||
|
||||
⚠️ **Do not read a rise between two `RUNS=3` runs as an improvement.** One
|
||||
sample is 33 points there, and the two weak-pointer runs above (33% / 33% and
|
||||
|
|
@ -452,7 +470,9 @@ break.
|
|||
hand-off has a `chain` case — a pointer between skills is otherwise measured
|
||||
by nothing at all, and a skill reached only that way scores a clean 100% on
|
||||
direct cases while the hand-off never fires. It does not count toward the
|
||||
floor in 4.
|
||||
floor in 4. **Then read the target's body as the caller arriving through that
|
||||
pointer** — the case proves the load, never that what loads is the half that
|
||||
caller needs (#2264).
|
||||
6. `npm run verify:skills` passes and the listing is under budget.
|
||||
7. `RUNS=5 npm run skills:eval` — the **whole** suite — is ≥80% on every
|
||||
first-move case, including the skills you did not touch, and the hand-off
|
||||
|
|
|
|||
Loading…
Reference in a new issue