project-stack-mappings.json had no `fastapi` entry, so /project-init
detected FastAPI projects as generic `python` and never resolved the
fastapi-patterns skill that already ships in the repo. api-design is
included for the same reason: FastAPI work is REST surface work.
Indicators match `fastapi` in requirements.txt or pyproject.toml,
following the existing django entry pattern. Data-only change; no
code reads this file, it is the lookup table /project-init consults.
Same class of bug as the session_id fix, one field over, found by code-reviewer
before merge: CLAUDE_HOOK_EVENT_NAME is never set for hook subprocesses either
(currently a harmless no-op since the dispatcher only ever calls this hook under
PostToolUse, but a real latent bug if that wiring ever changes). Also runs the
stdin-derived session_id through sanitizeSessionId(), matching the established
pattern already used in cost-tracker.js, ecc-context-monitor.js,
ecc-metrics-bridge.js, and gateguard-fact-force.js.
session-activity-tracker.js is correctly registered in
posttooluse-dispatcher.js's SYNC_HOOKS and fires on every tool call --
but buildActivityRow() only ever read the session id from
ECC_SESSION_ID/CLAUDE_SESSION_ID environment variables, which Claude
Code does not actually set for hook subprocesses. Every row silently
came back null and nothing was ever written to
~/.claude/metrics/tool-usage.jsonl, with no error anywhere.
The session id is available in the hook's own stdin JSON payload
(input.session_id) in real invocations. This prefers that, keeping the
env vars as a fallback. New test added; all 18 existing + new tests
pass.
Also adds a live Activity view to the capabilities dashboard
(scripts/dashboard-web.js) reading that same metrics file, so Skill
vs. Agent vs. raw-tool usage is visible in real time instead of
requiring a manual tail+jq -- verified live against a real session
before this commit. Documents the loopback-only, no-auth design and
the standard SSH-port-forwarding pattern for viewing it from another
device you control.
Adds /api/activity, reading ~/.claude/metrics/tool-usage.jsonl (the
session-activity-tracker hook wired in earlier), plus an Activity tab
polling it every 3s -- Skill/Agent calls visually distinct from raw
Bash/Edit/Write. Turns the tracker's log into an actual live proof
gate with a UI instead of a file someone has to tail and jq by hand.
Verified live: /api/activity already returns real entries from this
session before this commit was even pushed.
Adds a skill + matching delegated agent that draft Conventional
Commits-style messages by reading the repo's own git log norms
(prefix, scope, subject length) instead of assuming a fixed format.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Merge buildCommentMask into buildScanBoundaries so one pass owns comment mask and scan bounds.
Extract shell-scan primitives to scripts/hooks/lib/shell-scan.js; keep policy in the hook.
Use a sticky heredoc regex instead of copying the remaining input on every <<.
Replace the short-circuiting chain test with clean-leading sequences that exercise checkCommand advance.
Add quoted-executable non-leakage coverage for the scanEnd=length path.
Treat $'…' ANSI-C quoting as a quoted executable word.
Join unquoted backslash-newline for the flag scan and heredoc-line bound.
Co-Authored-By: grok (fleet lane) <noreply@fleet.local>
Co-authored-by: groy75 <groy75@users.noreply.github.com>
The scanner locked onto a `git commit` literal that sat inside a quoted
string of an OUTER command (a python heredoc's string, a VAR="..."
assignment, a printf JSON payload), then hasNoVerifyFlag() re-tokenized from
that inner offset with a fresh quote state. When quote parity flipped, the
"segment" ran past the enclosing string and picked up an unrelated later
`bash -n` / `sed -n` / `grep -n` as `commit -n`. Two legitimate commands were
blocked this way in one working session.
The first attempt at this PR (d5c428c) fixed the false positive by SKIPPING
`git` tokens judged to be in "data" context. That was the wrong lever: data
becomes executable the moment it is piped to a shell (`echo '...' | bash`,
`bash <<< '...'`, `cat <<EOF | bash`), and no regex list of shell invocations
is complete (`bash --noprofile -c` slipped). Three independent reviewers
(codex, Greptile, CodeRabbit) demonstrated executing bypasses; that commit is
squashed away here and none of its behaviour survives.
This version keeps every `git` candidate visible and keeps main's blocking
surface intact. A single O(n) preprocessing pass (buildScanBoundaries)
records, for each enclosed character, the closing quote or the current
heredoc-body line end. Subcommand discovery and flag tokenization are capped
at that boundary, so tokens after the enclosing data can never leak into a
candidate's argument list. Also: comment detection is a single mask pass
instead of a repeated prefix scan; heredoc parsing accepts quoted hyphenated
delimiters and CRLF terminators and preserves the `<<` vs `<<-` tab
distinction; quote-assembled words (`g''it`) are recognized.
Verified (all payloads run through the hook as JSON on stdin):
- red-list, 24 executing forms incl. every reviewer-reported bypass -> exit 2
on main AND here (CRLF heredoc trailer: 0 on main, 2 here);
- green-list, 16 leak-class false positives -> exit 2 on main, 0 here;
- tests/hooks/block-no-verify.test.js 84/84 (main's 25 retained verbatim,
d5c428c's 8 removed, 59 table-driven red/green cases added);
tests/hooks/cursor-block-no-verify.test.js 14/14;
- 204,807-byte command with 8,905 non-command `git` literals: 8.5 ms.
Co-Authored-By: Codex (GPT) <noreply@openai.com>
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EyuZY1LZNuShHiewExekLT
AGENTS.md requires functions under 50 lines; planOperations was 54 and mixed
source filtering, destination mapping, and operation construction. Extract the
per-path mapping into planSourcePathOperations and share the prefix test with
supportsCopilotSourcePath. planOperations is now 18 lines. No behaviour change.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
GitHub Copilot CLI discovers SKILL.md skills from .github/skills/ and custom
agents from .github/agents/, so ECC's catalog can be installed there without a
conversion step. This adds a managed-project install target that places both.
Scope is deliberately limited to agents and skills. ECC hooks are written
against Claude Code's event model, and commands rely on slash-command argument
substitution Copilot does not provide, so neither is installed.
Agent frontmatter goes through an allowlist transform rather than a passthrough:
Copilot resolves the session model from user configuration, so a Claude model id
makes it warn and fall back on every invocation; Claude tool names are not
Copilot tool names, and Copilot governs tool access per session via
--allow-tool/--deny-tool. Unknown keys added to ECC agents later therefore
cannot leak into the Copilot copy.
Target selection stays permissive in supportsModule so dependency-anchor modules
still resolve; planOperations is what narrows the install, matching the
convention used by the other adapters.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Reviewers without the CLI on PATH can reproduce the plain-clone count with the
npm-distributed build. The 39 + 3 split is reported by 1.0.82 from npm and by
1.0.83 installed locally, so it is not specific to one build or install method.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot CLI 1.0.83 lists all 42 plain-clone entries under Project skills with
source project and enabled true. Record that the three from .claude/commands/
are Claude command files surfaced as skills, and that their reported path is
the .claude/commands directory rather than a per-skill subdirectory.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The jq snippet filtered on source == "project", which returns 290 rather
than 286: the full catalog under .github/skills/ plus the .claude/commands/
entries and the one .agents/skills/ skill whose name does not collide with a
full-catalog entry. Filter on the path instead and explain the de-duplication.
Also document that Copilot registers a skill under its SKILL.md frontmatter
name rather than its directory name, so catalog folders whose directory and
frontmatter names differ still load.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot CLI discovers SKILL.md skills from .github/skills/, .agents/skills/, .claude/skills/, ~/.copilot/skills/ and ~/.agents/skills/, and custom agents from .github/agents/. ECC's existing skill and agent frontmatter is already compatible, so the README's 'no native skill discovery' and 'no subagent API' statements are out of date.
Because ECC ships a Codex skill subset in .agents/skills/, a plain clone already loads 42 ECC skills into Copilot CLI with no configuration. Placing skills/ at .github/skills/ loads all 286, all enabled.
Documentation only. No installer target, no hook porting, no new execution surface. Verified against Copilot CLI 1.0.83.
Keep npm and Yarn resolution policy aligned so both lockfile paths stay on the patched release.\n\nCo-authored-by: Svector-anu <svector-anu@users.noreply.github.com>
M Saad