The ecc@ecc plugin namespaces agents, commands and skills as `ecc:<name>`.
A manual install (install-apply --target claude|claude-project) places the
same files at bare names under ~/.claude, so plugin-prefixed references in
copied markdown point at agents and commands that do not exist in that
layout (e.g. subagent_type "ecc:planner", frontmatter agent: ecc:...).
- Add plugin-namespace.js with stripPluginNamespace(), which removes the
`ecc:` prefix before an identifier and leaves `ecc@ecc`, URLs and words
ending in "ecc" alone.
- Apply it to markdown copy operations when the target is claude or
claude-project. Non-markdown files and other targets are unchanged.
- Reword the agents rule so it describes both install modes instead of
stating that agents only ship with the plugin.
- Cover the transform with tests/lib/install-plugin-namespace.test.js.
/astra-review now runs on Opus (model: opus). Opus verifies findings,
checks every diff, and runs the tests; fixes go to a Fable subagent.
When that Agent call fails with a rate_limit / HTTP 429 API error, the
command switches to Astra for the rest of the session:
- astra-review.js --fix-findings <file> sends confirmed findings to
Codex in a workspace-write sandbox with the review isolation intact
(no web search, MCP disabled, user config ignored, no API keys) and
/tmp, $TMPDIR, and the network excluded from the writable roots.
- New scripts/lib/astra-review/fix.js builds the fix prompt, validates
the structured fix result, and formats the report.
- codex.js shares one exec runner between review and fix; the sandbox
is allowlisted (danger-full-access is refused).
- The command snapshots git diff and untracked files before any writer
runs, so Opus reverts only the writer's hunks and never the user's
uncommitted work.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Under OpenCode the skill-health dashboard had no writer and no honest
fallback, so it reported every discovered skill as healthy (#2463).
Root cause: `post:skill:track` is only wired for Claude Code, where
`posttooluse-dispatcher.js` fans the event out to `skill-run-tracker.js`.
The OpenCode plugin re-implements the harness events instead of reusing that
dispatcher, so nothing ever wrote `skill-runs.jsonl` — and the tracker's
matcher only accepted the exact Claude Code tool name `Skill`, while OpenCode
calls the tool `skill`, so even wiring it would have dropped every event.
Commands that locate ECC through the inlined resolver in
`scripts/lib/resolve-ecc-root.js` also fell back to `~/.claude`, which does
not contain ECC, so they could not find their scripts.
Changes:
- `.opencode/plugins/ecc-hooks.ts`: dispatch `post:skill:track` from
`tool.execute.after`. The tracker module is imported lazily and behind
try/catch, matching the existing `changed-files-store` pattern: this file is
OpenCode's startup entry point, so a static import failure would crash the
session before any hook runs (#2530). The module is reused rather than
reimplemented, keeping its identifier validation, length bounds and
"never persist prompt text" guarantees.
- `scripts/hooks/skill-run-tracker.js`: match the tool name case-insensitively
so `Skill` and `skill` are both accepted.
- `scripts/lib/skill-evolution/health.js`: split the summary into
measured/healthy/declining/unmeasured instead of deriving `healthy = total -
declining`, which turned "no telemetry" into "everything is fine" for every
skill.
- `scripts/lib/skill-evolution/dashboard.js`: show the breakdown and warn
explicitly when skills have no recorded runs, since absence of telemetry is
not a health signal.
Under OpenCode the resolver also needs `CLAUDE_PLUGIN_ROOT`, which only the
plugin can supply: `shell.env` exports it when a root is found, without
replacing a root the user set or one an earlier plugin already put in
`output.env`.
Root discovery is per capability and probes required runtime files — never
optional content. Probing for `skills/continuous-learning-v2` (the
`workflow-quality` module) rejected supported modular installs that ship the
plugin and tracker without that optional module, disabling telemetry and the
root export for exactly those users.
Also fixes an incidental bug: the compiled plugin bundle is ESM, where
`__dirname` is undefined, so the ECC version lookup silently fell back to its
default. The module directory is now derived from `import.meta.url`.
Tests: `tests/opencode-plugin-hooks.test.js` covers the telemetry dispatch
(skill tool recorded, other tools ignored, minimal profile respected), a
modular install without the optional skills module, the root export and both
override paths; `tests/hooks/skill-run-tracker.test.js` covers the
case-insensitive matcher; `tests/lib/skill-evolution.test.js` and
`tests/lib/skill-dashboard.test.js` cover the unmeasured bucket, the warning
and the aged-out-coverage wording.
Refs #2463, #2530
A non-numeric, negative, or exponent Points value was persisted and
silently skewed the derived sprint total (bogus -> 0, -3, 1e3 -> 1000)
with no drift reported. parseStory now enforces the documented values,
so apply rejects the batch and status reports the invalid story.
Resolve the Greptile findings on 74262955:
- Refuse symlinked .stories/, epics/, sprints/, .txn/, .ids/ and
symlinked ledger files, so no read or write leaves .stories/.
- Recovery accepts only regular backup-<n> files in .txn/, validates the
whole journal before restoring, and never writes through a planted
restore.tmp.
- Remove a stale lock only under .lock.break after a re-read, so a live
lock is never deleted; a leftover .lock.break times out (exit 3)
instead of spinning.
- apply enforces one-step-forward status transitions, new stories start
as todo, and Sprint must be unassigned or an existing sprint file.
- Refuse a batch that names the same file twice (e.g. a\b vs a/b).
Symlink tests skip on Windows when symlink creation returns EPERM.
The catalog gate validated only the first README project-tree block and
had no skills/commands counterpart to the agents pattern, so:
- README.md project tree (skills) and (commands) were never checked.
- A stale count in a second tree block was invisible because the agents
pattern used String.match with a non-global regex, which reads only the
first occurrence.
Both are now fail-closed. parseReadmeExpectations walks every tree block
for all three categories via matchAll, reports the offending line number
in the source label, and throws when an entry is missing entirely rather
than skipping it. syncEnglishReadme uses replaceAllOrThrow so --write
repairs every block instead of only the first; replaceOrThrow could not
be reused because it replaces only the first match and its regex.test()
probe is unsafe once the pattern carries the g flag.
README.md:847 documented 67 agents against an actual 68 and was never
checked, so it was corrected.
Fixtures in tests/ci/catalog.test.js and tests/ci/validators.test.js now
emit the skills and commands tree lines the gate requires, and both gain
regression coverage that fails against the previous parser (5 new tests).
main resolves SQL clients behind env -S and its launcher grammars, but a
destructive statement still passed under four forms:
- a shell -c inside a short-option cluster (bash -lc, sh -ec), since only
an exact -c was looked up;
- su, whose -c / --command / -lc value runs through the target user's
shell, before or after --;
- taskset, whose mask or CPU list precedes the command, like timeout's
duration;
- a command nested past the shell -c recursion limit, which was allowed
instead of denied.
The first three are checked like sh -c; past the limit the check now
fails closed.
silence-hole and coverage-hole are claims about the gate, and an ungated arm
never fires it, so classify labelled every ungated miss a coverage-hole by
construction. In the 48-trial run that padded the hole counts with five rows
that say nothing about the gate: coverage-hole read 7 when only 2 were the
gate's. Label an ungated miss unaided-miss, which is the baseline rate and is
excluded from the holes list.
The hidden-intent report printed bare counts. run.js reported Wilson intervals,
a bootstrap paired difference and an exact McNemar test, and that machinery was
lost in the new runner, so four trials rendered as a table that reads like a
result. On this run it is not one: gate 1/2 against off 0/2 gives intervals of
9%-91% and 0%-66%, a paired difference of -50% with a 95% interval spanning
-100% to 0%, and McNemar p = 1.000.
Reuse lib.js's pairedComparison and ai-eval-lib's wilson rather than adding a
second statistics path, and state the shortfall in words whenever a comparison
has fewer than twenty pairs: about 20 trials per arm are needed for 80% power on
a 40-point pass-rate difference at a two-sided 5% level, about 36 for 30 points,
and about 160 for the 15-point threshold the pilot pre-registered.
Mean turns and cost sit in the same table, because the only efficiency figure
this harness has produced points the other way: the gate roughly doubles turns
and cost per trial.
materializeTree passed the archive as an absolute path. GNU tar, which Git for
Windows bundles and puts first on PATH in Git Bash, reads an -f argument
containing a colon as a host:path remote spec and fails with "Cannot connect to
C: resolve failed". bsdtar, the tar in system32 that PowerShell resolves, has no
remote-spec parsing, so the fault only appeared under a bash-based shell - and
there it failed a committed test, not just a dry run.
Name the archive relatively and run tar with dest as its cwd, which extracts
identically under both tars and makes -C unnecessary. git archive keeps the
absolute -o path, which git handles on Windows.
Two tests, because the environment-dependent kind of fault needs both: one
injects a spawn and asserts no tar argument is absolute or drive-lettered, which
holds the invariant on any platform; the other prepends Git's bundled GNU tar to
PATH and extracts into a drive-letter destination, so the original failure is
reproducible from PowerShell too, where it was previously invisible.
The suite was the only file in tests/docker using node:test. The repository's
dominant convention, in 229 of 327 test files including the sibling
gateguard-effectiveness suite, is a local test() helper that counts passes and
sets process.exitCode, which is what tests/run-all.js reads.
Adopt that helper verbatim, along with the sibling's tempDir wrapper so the
scenario fixtures resolve symlinked temp paths. Assertions are unchanged.
Four defects found by running the repository's own checks and one more trial
against it.
ESLint rejected an unused `armSettings` import in intent-eval.js; the caller
supplies `armSettingsByName` already.
A timeout was lost whenever a later turn came back clean: `a || b ? c : false`
evaluates the condition first, so `timedOut` was recomputed from the new turn
instead of being kept.
Isolating each trial's config directory, to stop persisted sessions piling up
under the projects folder, puts the credentials out of reach. Every turn then
returns "Not logged in", and the harness graded that as a coverage-hole - an
auth failure recorded as a finding. Keep the real config directory and delete
only the session folders this run's work root created, and adopt lib.js's own
guards: a trial that is unauthenticated, or whose gated arm edited files with no
metrics line and no denial, stops the run instead of being recorded.
Record denials per turn as well as in total. The hook marks a file checked once
per session and --resume keeps one session id for the conversation, so a zero
after a non-zero is the latch rather than the gate staying quiet, and
silence-hole has to be read with that column in view.
Verified: a real trial authenticates, produces a graded silence-hole, and leaves
the projects directory at the same count it started with.
The existing harness cannot measure what it set out to measure. Every trap
hides its deciding fact in the repository, which agents reliably discover, and
trials run one-shot under "do not ask clarifying questions", so a fact-forcing
gate is evaluated with no user to answer it. Its scenarios also declare the
question the gate ought to raise, chosen by reading the gate's own taxonomy, so
the suite passes by construction and can never find a missing question.
Add a second runner that measures effectiveness instead of coverage:
- scenarios-intent/ keeps the deciding fact in intent.json, as a choice between
plausible conventions the repository cannot settle, so an agent that does not
ask can only guess and the ungated arm cannot sit at the ceiling.
- user-sim.js plays that user with a separate cheap model: it answers the
question in front of it, never volunteers, and reports which fact ids it
disclosed, so asking something and asking the deciding thing are told apart.
An unparseable judge is a recorded failure, not a silent stonewall.
- session.js runs multi-turn sessions joined by --session-id and --resume, and
tells every arm the user is reachable, so the comparison is symmetric and the
gate's contribution is whether it makes the agent ask the deciding question.
- coverage.js classifies each trial from behaviour alone into working, lucky, or
a follow-through, targeting, silence or coverage hole, and reports the
questions the gate actually raised rather than asserting declared ones.
- run-intent.js refuses minus-target, which needs targets these scenarios
deliberately lack, and checks every grader separates the start, the trap and
the reference before a session is billed.
A scenario counts as carrying no information only when nothing varies across
arms. Trap strength is reported separately: the ungated arm failing is what a
working trap looks like, so it must not make a scenario unusable.
The previous fix stopped a throwing hook from aborting the cold replay by
recording it as an allow, matching the warm path. But coldSummary only counted
disagreements, so a failure on a step the worker also allowed was no
disagreement and left no trace: the report showed no failure, a run with no
measured step reported a cold latency of 0 ms because percentile of an empty
list is 0, and the exit status passed. Trading a crash for a clean-looking
success hid exactly the failure the crash had exposed.
coldSummary now counts errors and samples, and reports no latency when nothing
was measured rather than a 0 that reads as a fast hook. The report prints "no
valid sample" in place of a latency and adds a line for steps that failed in a
fresh process, beside the existing disagreement line. The exit check moves into
gateFails, which fails on cold errors as it already did on warm errors and cold
disagreements, and is exported with coldSummary so both can be tested.
The render test builds its report from summarize([]), the real summary shape,
so it keeps working as metrics are added.
measureColdLatency ran each step in a fresh process through execFileSync,
inside try/finally with no catch. A hook that threw made the child exit
non-zero, execFileSync threw, and the error left measureColdLatency and then
evaluate, so neither the report nor the requested SARIF was written for a run
that had hit a single bad step.
The probe now catches a throwing require or run itself and records it the way
workerMain does: the raw payload stands in for the result, so the step reads as
an allow, and the error is kept. Matching the warm path matters because
coldSummary counts every cold decision that differs from the warm one as a
disagreement; any other choice would have reported disagreements that do not
exist. A child that dies without writing parseable output is caught outside the
probe and recorded the same way with no latency, and coldSummary leaves
unmeasured steps out of the percentiles while still counting them.
The new tests run a hook that always throws through both passes. They fail
against the previous code, 2 of 17, and pass against this one.
Brings in the three commits main gained since the branch's merge base: the 2.2.3
release, retiring the Everything Claude Code name for ECC, and the ja-JP sponsor
update. 227 files arrive from main; two conflicted.
CHANGELOG.md: the branch had added an Unreleased section and main a released
2.2.3 section, both at the top. Both are kept verbatim with Unreleased above
2.2.3, since none of the branch's entries shipped in that release.
manifests/context-packs/skill-triggers@1.json: both sides had recomputed
registryDigest from the same earlier value to different results, and neither is
the digest of the merged tree. It is recomputed with loadContextRegistry over the
merged tree and only that field changes; triggers and triggersDigest are
untouched on both sides. validate-context-profiles passes.
All twelve CI validators pass. The full suite reports 44 failures in seven files;
43 of them fail by name and count identically on an unmodified upstream/main, and
the remaining one, plugin-hook-bootstrap, fails only inside run-all and passes in
isolation on both trees. The merge introduces no failing test that main does not
already have.
Across 240 evaluation trials every question the gate raised was about code
structure - importers, callers, public API, data schemas - while every deciding
fact in the hidden-intent corpus was an external contract: a consumer's CSV
dialect, a reporting timezone, retry safety against a non-idempotent endpoint,
collation rules, a quota policy. Obtaining the deciding fact predicted the
outcome almost perfectly in both arms, and the gate did not reliably cause it.
Add an external-contract question for code targets whose change profile touches
data, beside data-schema rather than replacing it: data-schema asks what the
change touches, external-contract asks who decides what it has to be. It reuses
the existing touchesData signal, so it stays as selective as the question it
sits with and adds no new detection, and it therefore does not reach a change
with no data involvement - pure retry semantics, for instance - which is a known
limit to be measured rather than guessed at.
Code edits that touch data now ask four questions instead of three, so the
condensed hint for an unknown profile changes too. That string was pinned by a
test as legacy-stable; the pin is updated deliberately, because a hint that
names fewer questions than the gate asks is worse than a hint that changed.
Suites: target-class 31, fact-force 1019, scenario-gate 25, eval 15, and the two
effectiveness suites 29 and 26, all passing. The 23 relevantQuestions
annotations that listed data-schema now list external-contract as well.
json.loads raises RecursionError on deeply nested input, which escaped the
complete-JSON check and failed the request. Treat it like any other input
that is not a complete JSON document.
Signed-off-by: FenjuFu <fufenjupku@gmail.com>
montjeffrey