Skip to content

ccoalm/ccl-skills / skills/skill-extraction-workflow

skill-extraction-workflow

Flagged

复盘 / 沉淀 / 总结经验 / 补进技能 / 技能缺陷 / 流程优化 / would other teammates hit this / will we forget next time / teamwide recurrence / extract skill / post-mortem / 深度 review·审计技能仓库 / 对标外部技能包找 gap(benchmark superpowers·gstack 等) / 核查全局安装点(~/.config/opencode 等)旧快照是否遮蔽本仓技能 / 本仓(ccl-skills 等共享技能仓)OpenCode 项目配置·命令治理 → turn reusable lessons into durable CCL skills.

FileKindSize
SKILL.mdskill md111.0 KB
agents/openai.yamlconfig257 B
references/analysis-parse-fix-test-challenge-replay.mdmarkdown1.5 KB
references/attention-budget-ratchet.mdmarkdown13.7 KB
references/attribution-verification.mdmarkdown4.4 KB
references/bootstrap-slim-c3-obligation-table.mdmarkdown8.0 KB
references/correction-routing-map.mdmarkdown3.7 KB
references/coverage-exhaustion-traps.mdmarkdown19.6 KB
references/description-authoring.mdmarkdown21.1 KB
references/dual-track-review-gate.mdmarkdown110.6 KB
references/eval-routing.mdmarkdown22.2 KB
references/evidence-card-template.mdmarkdown3.2 KB
references/example-domain-preselect.mdmarkdown4.6 KB
references/external-practice-controls.mdmarkdown45.5 KB
references/extraction-lifecycle-handoff.mdmarkdown6.9 KB
references/extraction-quickstart.mdmarkdown20.2 KB
references/firing-point-placement.mdmarkdown21.3 KB
references/harness-patterns-and-eval.mdmarkdown42.6 KB
references/incident-postmortem-extraction.mdmarkdown28.8 KB
references/l0-l1-l2-routing.mdmarkdown10.3 KB
references/online-skill-review.mdmarkdown6.7 KB
references/parallel-stack-references-pattern.mdmarkdown35.9 KB
references/r0-leakage-audit.mdmarkdown19.9 KB
references/recurring-anti-patterns-checklist.mdmarkdown34.4 KB
references/resume-paused-delivery.mdmarkdown3.0 KB
references/review-feedback-mining.mdmarkdown5.7 KB
references/review-finding-standards.mdmarkdown7.3 KB
references/review-rubric.mdmarkdown3.1 KB
references/rule-consolidation.mdmarkdown38.2 KB
references/skill-listing-budget.mdmarkdown1.7 KB
references/source-register.mdmarkdown1.0 MB
references/source-to-skill-extraction.mdmarkdown109.3 KB
references/two-source-extraction-pattern.mdmarkdown19.4 KB
references/uiux-judgment-extraction.mdmarkdown19.1 KB
references/uiux-routing-map.mdmarkdown5.8 KB
references/validation-and-landing.mdmarkdown46.9 KB
scripts/AGENTS.mdmarkdown635 B
scripts/check-ccl-skills.shexecutablescript100.7 KB
scripts/check-contract-anchors.shexecutablescript5.6 KB
scripts/check-evidence-card-leak.shexecutablescript26.1 KB
scripts/check-mr-target-freshness.shexecutablescript9.4 KB
scripts/check-size-budget.shexecutablescript36.0 KB
scripts/check-sync-pointers.shexecutablescript22.6 KB
scripts/check_review_evidence_present.pyexecutablescript4.6 KB
scripts/contract-anchors.tsvother3.8 KB
scripts/entrypoint_form_census.pyexecutablescript7.1 KB
scripts/eval-golden-trace.rbexecutablescript10.0 KB
scripts/eval-health.rbexecutablescript14.6 KB
scripts/eval-routing-bank.rbexecutablescript30.4 KB
scripts/eval-routing.rbexecutablescript9.2 KB
scripts/extraction_review_gate.shexecutablescript1.8 KB
scripts/gate_receipt.pyexecutablescript24.7 KB
scripts/generic-r0-leak-scan.shexecutablescript12.4 KB
scripts/governing-chain-diff.pyexecutablescript12.8 KB
scripts/impact-chain-gate.rbexecutablescript124.4 KB
scripts/obligation-ledger.pyscript107.0 KB
scripts/reference-access-census.shexecutablescript8.3 KB
scripts/register-firing-path-resolution.rbexecutablescript48.4 KB
scripts/shared_git_surface_gate.pyexecutablescript38.0 KB
scripts/skill-behavior-eval.pyexecutablescript39.3 KB
scripts/skill-paired-eval.pyexecutablescript76.9 KB
scripts/source-register-lifecycle.rbscript1.5 KB
scripts/source-register-pending-status.rbscript2.0 KB
scripts/test_ai_coding_implementation_gates.shexecutablescript85.2 KB
scripts/test_antipattern_grep_panel.shexecutablescript3.2 KB
scripts/test_body_compliance_grading.shexecutablescript21.0 KB
scripts/test_check_ccl_impact_chain_refscripts.shexecutablescript145.1 KB
scripts/test_check_ccl_r0_status.shexecutablescript4.2 KB
scripts/test_check_ccl_register_pending_exclusion.shexecutablescript7.4 KB
scripts/test_check_ccl_regressions.shexecutablescript14.3 KB
scripts/test_check_ccl_route_drift.shexecutablescript22.6 KB
scripts/test_check_ccl_size_budget.shexecutablescript59.5 KB
scripts/test_check_ccl_skill_catalog.shexecutablescript23.3 KB
scripts/test_check_ccl_source_register_lifecycle.shexecutablescript6.1 KB
scripts/test_check_contract_anchors.shexecutablescript7.4 KB
scripts/test_check_mr_target_freshness.shexecutablescript12.8 KB
scripts/test_check_review_evidence_present.shexecutablescript4.3 KB
scripts/test_check_sync_pointers.shexecutablescript32.5 KB
scripts/test_ci_checkout_ref_binding.shexecutablescript8.2 KB
scripts/test_controlled_escalation_pins.shexecutablescript7.7 KB
scripts/test_entrypoint_domain_scan_terms.shexecutablescript8.8 KB
scripts/test_entrypoint_form_census.shexecutablescript6.3 KB
scripts/test_eval_routing_bank_grader_diagnostics.shexecutablescript20.6 KB
scripts/test_eval_routing_bank_resolution.shexecutablescript15.3 KB
scripts/test_eval_routing_bank_surface_binding.shexecutablescript6.9 KB
scripts/test_eval_routing_prose_target.shexecutablescript3.7 KB
scripts/test_eval_runtime.pyexecutablescript43.3 KB
scripts/test_extraction_review_gate.shexecutablescript15.2 KB
scripts/test_frozen_case_sanctity.shexecutablescript6.9 KB
scripts/test_frozen_case_sanctity_selfproof.shexecutablescript4.6 KB
scripts/test_gate_receipt.shexecutablescript17.7 KB
scripts/test_generic_r0_leak_scan.shexecutablescript6.4 KB
scripts/test_git_identity_predicate_gate.shscript13.8 KB
scripts/test_governing_chain_diff.shexecutablescript15.1 KB
scripts/test_impact_chain_gate_dateless_host.shexecutablescript6.8 KB
scripts/test_impact_chain_gate_verdict_differential.shexecutablescript20.8 KB
scripts/test_impact_chain_round_attribution.shexecutablescript32.5 KB
scripts/test_impact_chain_self_adjudication.shexecutablescript31.4 KB
scripts/test_impact_chain_source_refuted.shexecutablescript14.7 KB
scripts/test_liveness_predicate_gate.shexecutablescript14.5 KB
scripts/test_locale_independent_gates.shexecutablescript13.0 KB
scripts/test_obligation_ledger.shexecutablescript94.9 KB
scripts/test_obligation_ledger_repo_audit.shexecutablescript2.7 KB
scripts/test_pinned_phrase_mutation_walk.shexecutablescript6.5 KB
scripts/test_reference_access_census.shexecutablescript17.3 KB
scripts/test_register_firing_path_resolution.shexecutablescript41.0 KB
scripts/test_register_firing_path_wiring.shexecutablescript29.9 KB
scripts/test_regression_runner_lanes.shexecutablescript4.5 KB
scripts/test_regression_runner_registration.shexecutablescript1.5 KB
scripts/test_routing_bank_integrity.shexecutablescript15.5 KB
scripts/test_routing_pointer_integrity.shexecutablescript18.9 KB
scripts/test_shared_git_surface_gate.shexecutablescript76.8 KB
scripts/test_skill_paired_eval.pyscript60.0 KB
scripts/test_teardown_guard_pins.shexecutablescript17.0 KB
scripts/test_uiux_delivery_contract.shexecutablescript118.6 KB
scripts/test_uiux_loading_budget.shexecutablescript15.4 KB
scripts/test_validate_skill_credential_cwd.shexecutablescript2.6 KB
scripts/test_validate_skill_cross_refs.shexecutablescript4.8 KB
scripts/test_validate_skill_root_depth.shexecutablescript2.5 KB
scripts/validate-skill.shexecutablescript10.5 KB

scripts/check-mr-target-freshness.sh

View on GitHub
#!/usr/bin/env bash
# MR target-branch freshness gate.
#
# Failure class this blocks: a delivery branched off — and an MR targeted at — an
# integration branch that is no longer live. The observed instance: `dev` sat at
# 0 commits ahead of `main` and 279 behind, still carried the pre-slimming
# entrypoints, and ran an ADVISORY copy of the size gate that `main` runs as
# BLOCKING. An MR merged there is invisible on the real integration branch and
# silently bypasses whatever gates the default branch has since tightened.
#
# Predicate (decidable from git alone — no repo-policy knowledge, no API):
#   target is NOT the default branch
#   AND target is 0 commits AHEAD of the default branch   (carries nothing of its own)
#   AND target is >= STALE_BEHIND_MIN commits BEHIND it   (meaningfully abandoned)
#
# Both conjuncts matter. "Behind the default branch" alone would flag every
# legitimate long-lived target; the 0-ahead conjunct is what makes it precise:
#   - release/x.y while it carries cherry-picks or version bumps the default
#     lacks => ahead > 0 => never fires. A maintained release branch whose
#     patches have ALL been merged forward is legitimately 0-ahead and topology
#     cannot distinguish it from an abandoned mirror — that case is what
#     MR_TARGET_FRESHNESS_ALLOW exists for — a declared exemption, not an
#     authority check (see the enforcement boundary below).
#   - promote/* cut off the default — ahead-or-equal by construction, not behind
#     => never fires.
#   - a stacked MR targeting another feature branch — that branch has its own
#     commits => never fires.
#   - a dead mirror like `dev` — 0 ahead, far behind => fires.
# Scored against every remote branch in this repo when the gate was written:
# only the genuinely dead ones matched.
#
# Not an MR pipeline => no-op. Refs unresolvable, default branch undetermined,
# fork MR, or STALE_BEHIND_MIN malformed => fail CLOSED (exit 1): a gate that
# paints the job green while printing "not a pass" is the false-green this whole
# gate exists to prevent.
#
# ENFORCEMENT BOUNDARY — read before trusting this as a control.
# A merge-request pipeline runs the `.gitlab-ci.yml` of the SOURCE branch, and
# the merge request owns that file. Two consequences, both verified:
#   1. A branch cut from a stale base carries that base's CI config. `dev` has no
#      `check-mr-target-freshness` job at all, so an MR sourced from a dev-based
#      branch never creates this job — which is exactly the shape that motivated
#      the gate (MR !510). This script CANNOT catch that case.
#   2. EVERY input it reads can be overridden from candidate-owned YAML —
#      `CI_DEFAULT_BRANCH`, `MR_TARGET_FRESHNESS_ALLOW`, `STALE_BEHIND_MIN`,
#      `CI_MERGE_REQUEST_TARGET_BRANCH_NAME`, and both project-ID variables the
#      fork check reads. So the target and fork predicates are candidate-
#      influenced too, not just the allowlist and the threshold. No shell script
#      can establish that a variable came from a maintainer.
# So this is DEFENCE IN DEPTH, not an enforceable gate. What it does buy: a
# branch cut from a current base that is then mis-targeted at a dead branch gets
# caught, and the failure class becomes visible in CI output instead of being
# noticed only in review. Actually closing the class needs enforcement the merge
# request cannot replace — a project- or group-level required/compliance
# pipeline, or a server-side merge rule — which is a maintainer setting outside
# this repository's files. Do not describe this script as closing the class.
set -euo pipefail

root="${1:-$(pwd)}"
cd "$root"

target="${CI_MERGE_REQUEST_TARGET_BRANCH_NAME:-}"
if [ -z "$target" ]; then
  echo "mr_target_freshness_skipped: not a merge-request pipeline"
  exit 0
fi

default="${CI_DEFAULT_BRANCH:-}"
if [ -z "$default" ]; then
  default="$(git symbolic-ref --short refs/remotes/origin/HEAD 2>/dev/null | sed 's|^origin/||' || true)"
fi
if [ -z "$default" ]; then
  # Guessing "main" here would compare against a branch that may not be the
  # default at all. An undetermined default branch is an unevaluated gate.
  echo "mr_target_freshness_unevaluated: default branch is undetermined (CI_DEFAULT_BRANCH unset and origin/HEAD unresolvable) — this is NOT a pass" >&2
  exit 1
fi

# A fork MR's `origin` is the fork; the target branch lives in the upstream
# project, so same-named refs here would measure an unrelated branch.
src_pid="${CI_MERGE_REQUEST_SOURCE_PROJECT_ID:-}"
tgt_pid="${CI_MERGE_REQUEST_PROJECT_ID:-}"
if [ -n "$src_pid" ] || [ -n "$tgt_pid" ]; then
  if [ -z "$src_pid" ] || [ -z "$tgt_pid" ]; then
    # One present and one missing means the fork question cannot be answered;
    # treating that as same-project would be fail-open.
    echo "mr_target_freshness_unevaluated: incomplete merge-request project metadata (source='$src_pid' target='$tgt_pid') — cannot determine whether this is a fork; this is NOT a pass" >&2
    exit 1
  fi
  if [ "$src_pid" != "$tgt_pid" ]; then
    echo "mr_target_freshness_unevaluated: fork merge request (source project $src_pid != target project $tgt_pid) — origin is the fork, so its branches are not the target project's; this is NOT a pass" >&2
    exit 1
  fi
fi

if [ "$target" = "$default" ]; then
  echo "mr_target_freshness_ok: target=$target is the default branch"
  exit 0
fi

# A live integration target that is legitimately 0-ahead — e.g. a maintained
# release branch whose hotfixes have all been merged forward — cannot be told
# apart from an abandoned mirror by topology alone, so it needs a declared
# exemption. Intended to be set as a project CI variable; note that nothing
# here can enforce that, because candidate-owned YAML can set it too (see the
# enforcement boundary above). It is a declaration, not an authority check.
# `set -f` so a branch name is compared literally: an unquoted expansion would
# glob against the working tree and silently change allowlist membership.
set -f
for allowed in ${MR_TARGET_FRESHNESS_ALLOW:-}; do
  if [ "$target" = "$allowed" ]; then
    set +f
    echo "mr_target_freshness_ok: target=$target is in MR_TARGET_FRESHNESS_ALLOW (maintainer-declared live integration branch)"
    exit 0
  fi
done
set +f

stale_behind_min="${STALE_BEHIND_MIN:-50}"
case "$stale_behind_min" in
  '' | *[!0-9]*)
    echo "mr_target_freshness_misconfigured: STALE_BEHIND_MIN='$stale_behind_min' is not a non-negative integer — refusing to run rather than silently passing" >&2
    exit 1
    ;;
esac
# Digits alone are not enough: a value wider than a machine integer makes the
# `-ge` test abort with "integer expected", and because that abort happens in an
# `if` condition it is exempt from `set -e` and falls through to the ok path.
if [ "${#stale_behind_min}" -gt 9 ]; then
  echo "mr_target_freshness_misconfigured: STALE_BEHIND_MIN='$stale_behind_min' exceeds 9 digits — refusing rather than risking the numeric comparison aborting into a pass (not every such value overflows; 9 digits is a conservative bound, well above any real commit count)" >&2
  exit 1
fi

# ALWAYS force-fetch both refs. A reused runner checkout can carry a stale
# origin/<branch>, and measuring against historical topology is how a
# now-dead target would score as healthy.
# One fetch for both refspecs: two sequential fetches can observe two different
# moments. This narrows the window; it does not close it — a push to either
# branch after measurement can still leave a stale verdict, which is why this
# gate is a check on the MR as opened, not a merge-time invariant.
fetch_ok=1
git fetch --quiet --no-tags --force origin \
  "+refs/heads/$target:refs/remotes/origin/$target" \
  "+refs/heads/$default:refs/remotes/origin/$default" 2>/dev/null || fetch_ok=0

if [ "$fetch_ok" -eq 0 ] \
  || ! git rev-parse --verify --quiet "refs/remotes/origin/$target" >/dev/null \
  || ! git rev-parse --verify --quiet "refs/remotes/origin/$default" >/dev/null; then
  # Fail CLOSED. Exiting 0 here would paint the job green while the log says
  # "not a pass", which is the false-green this gate exists to prevent. This
  # repo clones with GIT_DEPTH: "0", so both refs are reachable in a healthy
  # pipeline; an unresolvable ref is a real problem, not routine noise.
  echo "mr_target_freshness_unevaluated: could not resolve origin/$target or origin/$default — this is NOT a pass, and the gate fails closed" >&2
  echo "  if the target genuinely lives in another project (fork MR), fetch it explicitly or declare it in MR_TARGET_FRESHNESS_ALLOW" >&2
  exit 1
fi

ahead="$(git rev-list --count "origin/$default..origin/$target")"
behind="$(git rev-list --count "origin/$target..origin/$default")"

echo "mr_target_freshness: target=$target default=$default ahead=$ahead behind=$behind threshold=$stale_behind_min"

if [ "$ahead" -eq 0 ] && [ "$behind" -ge "$stale_behind_min" ]; then
  {
    echo "mr_target_freshness_stale: target branch '$target' carries nothing the default branch '$default' lacks (ahead=0) and is $behind commits behind it"
    echo "  why this blocks: merging here does not reach '$default', and '$target' may still run older/weaker versions of the repo's gates than '$default' does"
    echo "  fix: retarget this MR to '$default', and rebase or re-cut the source branch off '$default'"
    echo "  if '$target' really is a live integration branch that is legitimately 0-ahead (all its patches merged forward), a maintainer declares it in MR_TARGET_FRESHNESS_ALLOW"
  } >&2
  exit 1
fi

echo "mr_target_freshness_ok"
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall