bigpowers 2.88.18 → 2.89.1

package/.gemini/extensions/bigpowers/gemini-extension.json

@@ -1,7 +1,7 @@
 {
   "name": "bigpowers",
-  "version": "2.88.18",
-  "description": "81 skills — 73 agent skills synthesizing 17 years of software engineering discipline into a prescriptive methodology for solo developers",
+  "version": "2.89.1",
+  "description": "80 skills — 73 agent skills synthesizing 17 years of software engineering discipline into a prescriptive methodology for solo developers",
   "keywords": [
     "gemini-extension"
   ],

package/.pi/package.json

@@ -1,7 +1,7 @@
 {
   "name": "bigpowers",
-  "version": "2.88.18",
-  "description": "81 skills — 73 agent skills synthesizing 17 years of software engineering discipline into a prescriptive methodology for solo developers",
+  "version": "2.89.1",
+  "description": "80 skills — 73 agent skills synthesizing 17 years of software engineering discipline into a prescriptive methodology for solo developers",
   "keywords": [
     "pi-package"
   ],

package/CHANGELOG.md

@@ -1,3 +1,17 @@
+## [2.89.1](https://github.com/danielvm-git/bigpowers/compare/v2.89.0...v2.89.1) (2026-11-03)
+
+
+### Bug Fixes
+
+* **scripts:** remove pipefail SIGPIPE races from git and long-output pipelines ([#186](https://github.com/danielvm-git/bigpowers/issues/186)) ([f55cbb7](https://github.com/danielvm-git/bigpowers/commit/f55cbb7d8883dd0af4777ac2467d5b550fde23db))
+
+# [2.89.0](https://github.com/danielvm-git/bigpowers/compare/v2.88.18...v2.89.0) (2026-11-02)
+
+
+### Features
+
+* **skills:** remove context7-mcp skill for provider-agnostic catalog ([#180](https://github.com/danielvm-git/bigpowers/issues/180)) ([514826b](https://github.com/danielvm-git/bigpowers/commit/514826b9c5c003aad45349e9cf54b2f09976a7ce))
+
 ## [2.88.18](https://github.com/danielvm-git/bigpowers/compare/v2.88.17...v2.88.18) (2026-11-02)
 
 

package/CLAUDE.md

@@ -173,7 +173,7 @@ Before any task, run this sequence — not optional:
 - Search with `bts find` before opening files to locate a symbol or pattern.
 - Pipe anything > 200 lines through `bts compress` before adding to context.
 - Run `bts map` when asked for a repo overview.
-- Use `bts docs <lib>` before answering questions about library APIs. Doc fetches use `scripts/lib/doc-fetch-cache.sh` (ETag-revalidated, 300s TTL — see `context7-mcp` skill, e45s20).
+- Use `bts docs <lib>` before answering questions about library APIs. Doc fetches use `scripts/lib/doc-fetch-cache.sh` (ETag-revalidated, 300s TTL).
 - If a tool is missing, say so and run `bts doctor` — do not silently substitute.
 
 <!-- BEGIN rtk-pretooluse-hook (e45s16 — mechanical PreToolUse backstop; remove block to disable) -->

package/CONVENTIONS.md

@@ -380,7 +380,6 @@ name must return < 5 results across the repo.
 | `visual-dashboard` | adjective-noun | `view-dashboard` implies read-only; `show-dashboard` collides with `show` verbs |
 | `deploy` | single verb | Well-known DevOps single-word concept; renames like `deploy-app` or `deploy-service` are redundant since deploy always targets an application |
 | `grill-with-docs` | three-word verb-prep-noun | Variant of `grill-me`; the `-with-docs` suffix is the distinguishing feature — collapsing it would hide the doc-grounded contract |
-| `context7-mcp` | noun-noun (product + protocol) | Names the external service and protocol it fronts; a verb-noun rename (`fetch-docs`) would collide with generic doc-fetch skills and hide which provider is used |
 | `generate-allure-report` | three-word verb-noun-noun | `generate-report` is ambiguous across report types; `allure` names the specific output format consumers integrate with |
 | `simple-english` | adjective-noun | Names the recognizable STE brand users search for (`simplify-english` / `write-plain` hide intent and break searchability for users coming from the upstream skill) |
 

package/README.md

@@ -2,7 +2,7 @@
 
 ![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)
 ![npm version](https://img.shields.io/npm/v/bigpowers.svg)
-![Skills](https://img.shields.io/badge/skills-81-brightgreen.svg)
+![Skills](https://img.shields.io/badge/skills-80-brightgreen.svg)
 
 **Agent skills synthesizing 17 years of software engineering discipline — from Clean Code to AI-native architecture — into a single, prescriptive methodology for solo developers.**
 

package/SKILL-INDEX.md

@@ -3,7 +3,7 @@
 > **DO NOT EDIT** — This file is auto-generated by `scripts/generate-skill-index.sh`.
 > Edit `SKILL.md` source files or `skills-lock.json` instead. Run `bash scripts/sync-skills.sh` to regenerate.
 
-**Skills:** 81
+**Skills:** 80
 
 ---
 
@@ -15,10 +15,10 @@
 | Design | 7 | `deepen-architecture, define-language, define-success, design-interface, grill-me, grill-with-docs, model-domain` |
 | Plan | 9 | `assess-impact, change-request, plan-refactor, plan-release, plan-work, run-planning, scope-work, seed-conventions, slice-tasks` |
 | Build | 17 | `align-grid, build-epic, craft-skill, deploy, develop-tdd, execute-plan, extract-design, guard-git, hook-commits, kickoff-branch, orchestrate-project, publish-package, quick-fix, setup-environment, spike-prototype, wire-ci, wire-observability` |
-| Verify | 20 | `audit-code, context7-mcp, diagnose-root, diagnose-stall, enforce-first, fix-bug, gate-trace, generate-allure-report, inspect-quality, investigate-bug, plan-tests, request-review, respond-review, run-evals, security-review, smoke-test, trace-requirement, validate-contracts, validate-fix, verify-work` |
+| Verify | 19 | `audit-code, diagnose-root, diagnose-stall, enforce-first, fix-bug, gate-trace, generate-allure-report, inspect-quality, investigate-bug, plan-tests, request-review, respond-review, run-evals, security-review, smoke-test, trace-requirement, validate-contracts, validate-fix, verify-work` |
 | Release | 2 | `commit-message, release-branch` |
 | Sustain | 18 | `compose-workflow, delegate-task, dispatch-agents, edit-document, evolve-skill, harden-vps, maintain-wiki, migrate-spec, organize-workspace, reset-baseline, run-benchmark, session-state, simple-english, simulate-agents, stocktake-skills, terse-mode, visual-dashboard, write-document` |
-| **TOTAL** | **81** | |
+| **TOTAL** | **80** | |
 
 ---
 
@@ -68,47 +68,46 @@
 | 40 | Build | `wire-ci` | "CI pipeline setup with bundled, forge-neutral templates and local validation. D | ✅ Active |
 | 41 | Build | `wire-observability` | Add structured JSON logging, observability commands, and idempotent setup script | ✅ Active |
 | 42 | Verify | `audit-code` | Self-review checklist for the coding agent to run before dispatching a reviewer. | ✅ Active |
-| 43 | Verify | `context7-mcp` | Fetch current library docs via Context7 MCP instead of training data. Use when u | ✅ Active |
-| 44 | Verify | `diagnose-root` | Run 4-phase root cause analysis — reproduce, isolate, hypothesize, verify. Use | ✅ Active |
-| 45 | Verify | `diagnose-stall` | Diagnose why agent orchestration stopped producing progress — silent stalls in | ✅ Active |
-| 46 | Verify | `enforce-first` | Apply the F.I.R.S.T test quality rubric (per CONVENTIONS.md §Tests) to a test s | ✅ Active |
-| 47 | Verify | `fix-bug` | Bug fix orchestrator — active_flow fix_bug; reads specs/bugs/BUG-*.md; chains  | ✅ Active |
-| 48 | Verify | `gate-trace` | "Deterministic traceability quality gate — reads coverage matrix + blind-spot  | ✅ Active |
-| 49 | Verify | `generate-allure-report` | "Generate Allure-ready reports from bigpowers YAML metadata. Reads execution-sta | ✅ Active |
-| 50 | Verify | `inspect-quality` | Interactive QA session where user reports bugs or issues conversationally, and t | ✅ Active |
-| 51 | Verify | `investigate-bug` | Investigate a bug or issue by exploring the codebase to find root cause, then wr | ✅ Active |
-| 52 | Verify | `plan-tests` | "Design a risk-scaled test architecture for an epic before implementation begins | ✅ Active |
-| 53 | Verify | `request-review` | Dispatch a fresh reviewer agent with a clean context to critique the code after  | ✅ Active |
-| 54 | Verify | `respond-review` | Act on a reviewer agent's feedback systematically — categorize findings, apply | ✅ Active |
-| 55 | Verify | `run-evals` | Eval-Driven Development — define capability and regression evals before buildi | ✅ Active |
-| 56 | Verify | `security-review` | > | ✅ Active |
-| 57 | Verify | `smoke-test` | "Post-deploy health-check against a live URL. Validates HTTP status, response co | ✅ Active |
-| 58 | Verify | `trace-requirement` | Link story IDs from specs/release-plan.yaml + epic capsule directories to the im | ✅ Active |
-| 59 | Verify | `validate-contracts` | "Assert data shape consistency across system boundaries — live API responses a | ✅ Active |
-| 60 | Verify | `validate-fix` | Prove a fix works before declaring done — re-run the failing test, run the ful | ✅ Active |
-| 61 | Verify | `verify-work` | Multi-phase UAT gate — cold-start smoke, build, typecheck, lint, tests, step-b | ✅ Active |
-| 62 | Release | `commit-message` | Reviews working-tree changes, then drafts a Conventional Commits title/body and  | ✅ Active |
-| 63 | Release | `release-branch` | Make the merge/PR/keep/discard decision for a feature branch, verify coverage ga | ✅ Active |
-| 64 | Sustain | `compose-workflow` | Chain multiple bigpowers skills into a custom workflow recipe saved in specs/. U | ✅ Active |
-| 65 | Sustain | `delegate-task` | Delegate one complex task to a single subagent, review its work in two stages be | ✅ Active |
-| 66 | Sustain | `dispatch-agents` | Dispatch multiple subagents in parallel on independent tasks. No waiting between | ✅ Active |
-| 67 | Sustain | `edit-document` | Edit and improve documents by restructuring sections, improving clarity, and tig | ✅ Active |
-| 68 | Sustain | `evolve-skill` | Benchmark-gated skill evolution — consume bigpowers-benchmark report, propose  | ✅ Active |
-| 69 | Sustain | `harden-vps` | "Harden a production Linux VPS for your application across three layers — appl | ✅ Active |
-| 70 | Sustain | `maintain-wiki` | "Agent-maintained OKF wiki — INGEST source docs, LINT for issues, QUERY across | ✅ Active |
-| 71 | Sustain | `migrate-spec` | Detect GSD, spec-kit, or BMAD spec artifacts and transform them into bigpowers Y | ✅ Active |
-| 72 | Sustain | `organize-workspace` | Scans the active workspace for disposable artifacts—logs, caches, stale build  | ✅ Active |
-| 73 | Sustain | `reset-baseline` | Restore the project to a known clean state between agent runs or experiments. Us | ✅ Active |
-| 74 | Sustain | `run-benchmark` | Run skill quality benchmarks from specs/benchmarks/ definitions — N-run with/w | ✅ Active |
-| 75 | Sustain | `session-state` | Track implementation decisions and progress in specs/state.yaml to prevent conte | ✅ Active |
-| 76 | Sustain | `simple-english` | Write or rewrite technical text with the rules of ASD-STE100 Simplified Technica | ✅ Active |
-| 77 | Sustain | `simulate-agents` | Run Mock User and Auditor agents against a feature in fresh contexts before huma | ✅ Active |
-| 78 | Sustain | `stocktake-skills` | Sequential subagent batch audit of the bigpowers skill catalog — Quick Scan (c | ✅ Active |
-| 79 | Sustain | `terse-mode` | Fallback ultra-compressed communication mode. Cuts token usage ~75% by dropping  | ✅ Active |
-| 80 | Sustain | `visual-dashboard` | Start a browser-based dashboard that visualizes architecture, implementation pla | ✅ Active |
-| 81 | Sustain | `write-document` | Write, organize, and sync high-integrity technical documents using the BMAD meth | ✅ Active |
-
-**Total: 81 active skills.**
+| 43 | Verify | `diagnose-root` | Run 4-phase root cause analysis — reproduce, isolate, hypothesize, verify. Use | ✅ Active |
+| 44 | Verify | `diagnose-stall` | Diagnose why agent orchestration stopped producing progress — silent stalls in | ✅ Active |
+| 45 | Verify | `enforce-first` | Apply the F.I.R.S.T test quality rubric (per CONVENTIONS.md §Tests) to a test s | ✅ Active |
+| 46 | Verify | `fix-bug` | Bug fix orchestrator — active_flow fix_bug; reads specs/bugs/BUG-*.md; chains  | ✅ Active |
+| 47 | Verify | `gate-trace` | "Deterministic traceability quality gate — reads coverage matrix + blind-spot  | ✅ Active |
+| 48 | Verify | `generate-allure-report` | "Generate Allure-ready reports from bigpowers YAML metadata. Reads execution-sta | ✅ Active |
+| 49 | Verify | `inspect-quality` | Interactive QA session where user reports bugs or issues conversationally, and t | ✅ Active |
+| 50 | Verify | `investigate-bug` | Investigate a bug or issue by exploring the codebase to find root cause, then wr | ✅ Active |
+| 51 | Verify | `plan-tests` | "Design a risk-scaled test architecture for an epic before implementation begins | ✅ Active |
+| 52 | Verify | `request-review` | Dispatch a fresh reviewer agent with a clean context to critique the code after  | ✅ Active |
+| 53 | Verify | `respond-review` | Act on a reviewer agent's feedback systematically — categorize findings, apply | ✅ Active |
+| 54 | Verify | `run-evals` | Eval-Driven Development — define capability and regression evals before buildi | ✅ Active |
+| 55 | Verify | `security-review` | > | ✅ Active |
+| 56 | Verify | `smoke-test` | "Post-deploy health-check against a live URL. Validates HTTP status, response co | ✅ Active |
+| 57 | Verify | `trace-requirement` | Link story IDs from specs/release-plan.yaml + epic capsule directories to the im | ✅ Active |
+| 58 | Verify | `validate-contracts` | "Assert data shape consistency across system boundaries — live API responses a | ✅ Active |
+| 59 | Verify | `validate-fix` | Prove a fix works before declaring done — re-run the failing test, run the ful | ✅ Active |
+| 60 | Verify | `verify-work` | Multi-phase UAT gate — cold-start smoke, build, typecheck, lint, tests, step-b | ✅ Active |
+| 61 | Release | `commit-message` | Reviews working-tree changes, then drafts a Conventional Commits title/body and  | ✅ Active |
+| 62 | Release | `release-branch` | Make the merge/PR/keep/discard decision for a feature branch, verify coverage ga | ✅ Active |
+| 63 | Sustain | `compose-workflow` | Chain multiple bigpowers skills into a custom workflow recipe saved in specs/. U | ✅ Active |
+| 64 | Sustain | `delegate-task` | Delegate one complex task to a single subagent, review its work in two stages be | ✅ Active |
+| 65 | Sustain | `dispatch-agents` | Dispatch multiple subagents in parallel on independent tasks. No waiting between | ✅ Active |
+| 66 | Sustain | `edit-document` | Edit and improve documents by restructuring sections, improving clarity, and tig | ✅ Active |
+| 67 | Sustain | `evolve-skill` | Benchmark-gated skill evolution — consume bigpowers-benchmark report, propose  | ✅ Active |
+| 68 | Sustain | `harden-vps` | "Harden a production Linux VPS for your application across three layers — appl | ✅ Active |
+| 69 | Sustain | `maintain-wiki` | "Agent-maintained OKF wiki — INGEST source docs, LINT for issues, QUERY across | ✅ Active |
+| 70 | Sustain | `migrate-spec` | Detect GSD, spec-kit, or BMAD spec artifacts and transform them into bigpowers Y | ✅ Active |
+| 71 | Sustain | `organize-workspace` | Scans the active workspace for disposable artifacts—logs, caches, stale build  | ✅ Active |
+| 72 | Sustain | `reset-baseline` | Restore the project to a known clean state between agent runs or experiments. Us | ✅ Active |
+| 73 | Sustain | `run-benchmark` | Run skill quality benchmarks from specs/benchmarks/ definitions — N-run with/w | ✅ Active |
+| 74 | Sustain | `session-state` | Track implementation decisions and progress in specs/state.yaml to prevent conte | ✅ Active |
+| 75 | Sustain | `simple-english` | Write or rewrite technical text with the rules of ASD-STE100 Simplified Technica | ✅ Active |
+| 76 | Sustain | `simulate-agents` | Run Mock User and Auditor agents against a feature in fresh contexts before huma | ✅ Active |
+| 77 | Sustain | `stocktake-skills` | Sequential subagent batch audit of the bigpowers skill catalog — Quick Scan (c | ✅ Active |
+| 78 | Sustain | `terse-mode` | Fallback ultra-compressed communication mode. Cuts token usage ~75% by dropping  | ✅ Active |
+| 79 | Sustain | `visual-dashboard` | Start a browser-based dashboard that visualizes architecture, implementation pla | ✅ Active |
+| 80 | Sustain | `write-document` | Write, organize, and sync high-integrity technical documents using the BMAD meth | ✅ Active |
+
+**Total: 80 active skills.**
 
 ---
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "bigpowers",
-  "version": "2.88.18",
+  "version": "2.89.1",
   "description": "73 agent skills synthesizing 17 years of software engineering discipline into a prescriptive methodology for solo developers",
   "main": "index.js",
   "scripts": {

package/scripts/bp-churn-rank.sh

@@ -45,5 +45,6 @@ git log --since="$SINCE" --name-only --pretty=format: $PATHSPEC \
   | sort \
   | uniq -c \
   | sort -rn \
-  | head -n "$LIMIT" \
-  | awk '{printf "%4d  %s\n", $1, $2}'
+  | awk -v limit="$LIMIT" 'NR <= limit {printf "%4d  %s\n", $1, $2}'
+# awk reads to the end instead of `head -n`, which closed the pipe early and
+# made `sort` die with SIGPIPE under pipefail (BUG-2026-11-02-pipefail-pipe-race).

package/scripts/check-spec-version-gap.sh

@@ -81,7 +81,8 @@ yaml_value_set() {
 }
 
 has_dirty_specs() {
-  [[ -d "$SPECS" ]] && git -C "$PROJECT_DIR" status --porcelain -- specs/ 2>/dev/null | grep -q .
+  # Test the captured output, not `git status | grep -q .` — a pipefail race (BUG-2026-11-02-pipefail-pipe-race).
+  [[ -d "$SPECS" ]] && [[ -n "$(git -C "$PROJECT_DIR" status --porcelain -- specs/ 2>/dev/null)" ]]
 }
 
 is_blocking_active_flow() {

package/scripts/generate-skill-index.sh

@@ -44,7 +44,7 @@ phase_of() {
     # Build
     develop-tdd|kickoff-branch|execute-plan|build-epic|spike-prototype|craft-skill|quick-fix|setup-environment|wire-observability|wire-ci|publish-package|align-grid|orchestrate-project|guard-git|hook-commits|deploy|smoke-test|validate-contracts|extract-design) echo "Build" ;;
     # Verify
-    verify-work|validate-fix|audit-code|enforce-first|run-evals|investigate-bug|diagnose-root|fix-bug|inspect-quality|request-review|respond-review|trace-requirement|security-review|context7-mcp|diagnose-stall|gate-trace|generate-allure-report|plan-tests) echo "Verify" ;;
+    verify-work|validate-fix|audit-code|enforce-first|run-evals|investigate-bug|diagnose-root|fix-bug|inspect-quality|request-review|respond-review|trace-requirement|security-review|diagnose-stall|gate-trace|generate-allure-report|plan-tests) echo "Verify" ;;
     # Release
     release-branch|commit-message) echo "Release" ;;
     # Sustain

package/scripts/land-branch.sh

@@ -86,8 +86,11 @@ if ! git show-ref --verify --quiet "refs/heads/$FEATURE_BRANCH"; then
   land_branch_deny "Feature branch '$FEATURE_BRANCH' does not exist"
 fi
 
-# Scan all commits in feature branch for Co-authored-by: footers
-if git log "$DEFAULT_BRANCH..$FEATURE_BRANCH" --format="%B" 2>/dev/null | grep -qiE '^co[- ]authored[- ]by:'; then
+# Scan all commits in feature branch for Co-authored-by: footers.
+# Capture first: `git log | grep -q` under pipefail fails at random when grep
+# exits before git finishes writing (SIGPIPE) — BUG-2026-11-02-pipefail-pipe-race.
+FEATURE_BODIES="$(git log "$DEFAULT_BRANCH..$FEATURE_BRANCH" --format="%B" 2>/dev/null || true)"
+if grep -qiE '^co[- ]authored[- ]by:' <<<"$FEATURE_BODIES"; then
   land_branch_deny "Feature branch '$FEATURE_BRANCH' contains Co-authored-by: footer(s). Amend commits to remove all AI agent attribution before landing."
 fi
 
@@ -194,7 +197,8 @@ fi
 
 # Worktree cleanup
 WORKTREE_PATH="../$FEATURE_BRANCH"
-if git worktree list --porcelain 2>/dev/null | grep -q "^worktree $WORKTREE_PATH$"; then
+WORKTREES="$(git worktree list --porcelain 2>/dev/null || true)"
+if grep -q "^worktree $WORKTREE_PATH$" <<<"$WORKTREES"; then
   echo "==> Removing worktree $WORKTREE_PATH"
   git worktree remove "$WORKTREE_PATH" 2>/dev/null || git worktree remove -f "$WORKTREE_PATH"
 fi

package/scripts/lib/doc-fetch-cache.sh

@@ -1,6 +1,6 @@
 #!/usr/bin/env bash
 # story: e45s20
-# doc-fetch-cache.sh — ETag-revalidated fetch cache for Context7 / bts docs queries.
+# doc-fetch-cache.sh — ETag-revalidated fetch cache for bts docs queries.
 # Usage:
 #   bash scripts/lib/doc-fetch-cache.sh get <cache-key>     # print cached body or empty
 #   bash scripts/lib/doc-fetch-cache.sh put <key> <etag> <body-file>

package/scripts/lib/golden-suite-gates.sh

@@ -79,6 +79,8 @@ GOLDEN_GATES=(
   "specs-parse:bash scripts/validate-specs-yaml.sh:false"
   # The runner itself: failing output is shown, skips are named and reasoned.
   "golden-gate-runner:bash scripts/test-golden-gate-runner.sh:false"
+  # Pipefail SIGPIPE races: fix shapes hold, git output never feeds grep -q/head.
+  "pipefail-pipe-race:bash scripts/test-pipefail-pipe-race.sh:false"
   # One pipeline, release needs: every gate, pinned runner and actions.
   "ci-pipeline-contract-selftest:bash scripts/test-ci-pipeline-contract.sh --self-test:false"
   "ci-pipeline-contract:bash scripts/test-ci-pipeline-contract.sh:false"

package/scripts/lib/parallel-review-worktrees.sh

@@ -11,7 +11,9 @@ cd "$REPO_ROOT"
 BASE=".bigpowers/worktrees/review-${CHECK}"
 mkdir -p "$(dirname "$BASE")"
 
-if git worktree list | grep -q "$BASE"; then
+# Capture first: `git worktree list | grep -q` is a pipefail race (BUG-2026-11-02-pipefail-pipe-race).
+WORKTREES="$(git worktree list)"
+if grep -q "$BASE" <<<"$WORKTREES"; then
   git worktree remove --force "$BASE" 2>/dev/null || true
 fi
 

package/scripts/lib/sync-post.sh

@@ -73,7 +73,7 @@ sync_post_run() {
     bash "$REPO_ROOT/scripts/build-skill-index.sh" || true
   fi
 
-  local CURSOR_KEEP="context7.mdc" mdc mdc_base
+  local CURSOR_KEEP="" mdc mdc_base
   for mdc in "$CURSOR_RULES"/*.mdc; do
     [[ -e "$mdc" ]] || continue
     mdc_base=$(basename "$mdc")

package/scripts/run-smoke.sh

@@ -46,7 +46,8 @@ run_single_url_check() {
 
   # Assert content signal
   if [ -n "$content_signal" ]; then
-    if echo "$body" | grep -qiE "$content_signal"; then
+    # Here-string, not `echo "$body" | grep -q`: a large body races pipefail (BUG-2026-11-02-pipefail-pipe-race).
+    if grep -qiE "$content_signal" <<<"$body"; then
       echo "  PASS: body matches \"$content_signal\""
     else
       echo "  FAIL: body does not match \"$content_signal\""
@@ -165,7 +166,7 @@ run_parsed_check() {
 
   # Assert content signal
   if [ -n "$content_signal" ]; then
-    if echo "$body" | grep -qiE "$content_signal"; then
+    if grep -qiE "$content_signal" <<<"$body"; then
       echo "  PASS: body matches \"$content_signal\""
       checks_passed=$((checks_passed + 1))
     else

package/scripts/test-pipefail-pipe-race.sh

@@ -0,0 +1,68 @@
+#!/usr/bin/env bash
+# story: BUG-2026-11-02-pipefail-pipe-race
+# test-pipefail-pipe-race.sh — regression guard for SIGPIPE races under pipefail.
+#
+# `producer | grep -q` (or `| head`) fails at random under `set -o pipefail`:
+# the reader exits early, the producer dies with SIGPIPE (141), and pipefail
+# reports the pipeline as failed — a found match reads as "not found".
+#
+#   (a) anti-vacuity: the racy shape really fails on this host
+#   (b) each fix shape used in the repo passes RUNS times with a long producer
+#   (c) static guard: no pipefail script pipes git output into an early-exit reader
+#
+# Usage: bash scripts/test-pipefail-pipe-race.sh
+
+set -euo pipefail
+source "$(dirname "${BASH_SOURCE[0]}")/lib/skill-common.sh"
+resolve_repo_root
+cd "$REPO_ROOT"
+
+source scripts/lib/test-assertions.sh
+TA_PASS=0
+TA_FAIL=0
+RUNS=20
+
+# ~2 MB of output: far past any pipe buffer, so an early reader always wins.
+long_producer() { seq 1 300000; }
+
+# Runs "$@" RUNS times; prints how many runs returned non-zero.
+count_failures() {
+  local i fails=0
+  for ((i = 0; i < RUNS; i++)); do
+    "$@" || fails=$((fails + 1))
+  done
+  echo "$fails"
+}
+
+racy_grep_q()      { long_producer | grep -q '^1$'; }
+fixed_capture()    { local out; out="$(long_producer)"; grep -q '^1$' <<<"$out"; }
+fixed_nonempty()   { [[ -n "$(long_producer)" ]]; }
+fixed_awk_limit()  { long_producer | awk -v limit=3 'NR <= limit' >/dev/null; }
+fixed_sed_range()  { long_producer | sed -n '1,5p' >/dev/null; }
+
+echo "=== test-pipefail-pipe-race.sh ==="
+
+fails=$(count_failures racy_grep_q)
+[[ "$fails" -gt 0 ]] \
+  && ta_pass "(a) racy 'producer | grep -q' fails under pipefail ($fails/$RUNS)" \
+  || ta_fail "(a) racy shape never failed — test cannot detect the race on this host"
+
+for fix in fixed_capture fixed_nonempty fixed_awk_limit fixed_sed_range; do
+  fails=$(count_failures "$fix")
+  [[ "$fails" -eq 0 ]] \
+    && ta_pass "(b) $fix: 0/$RUNS failures" \
+    || ta_fail "(b) $fix: $fails/$RUNS failures"
+done
+
+# run-skill-verify.sh holds `git diff ... | head -20` as a fixture string of
+# fail-open verify directives, not as a command it runs.
+hits=$(grep -rlE 'pipefail' scripts hooks --include='*.sh' \
+  | grep -v '^scripts/run-skill-verify\.sh$' \
+  | xargs grep -nE 'git [a-z-]+[^|]*\|[[:space:]]*(grep -[a-zA-Z]*[qm]|head)' \
+  | grep -vE '^[^:]+:[0-9]+:[[:space:]]*#' || true)
+[[ -z "$hits" ]] \
+  && ta_pass "(c) no pipefail script pipes git output into grep -q/-m or head" \
+  || ta_fail "(c) git output piped into an early-exit reader:"$'\n'"$hits"
+
+echo "$TA_PASS passed, $TA_FAIL failed"
+[[ "$TA_FAIL" -eq 0 ]]

package/scripts/validate-doctrine.sh

@@ -55,7 +55,7 @@ else
   grep -rE "PLAN\.md|STATE\.md|PROJECT\.md|CONTEXT\.md|SUMMARY\.md|VERIFICATION\.md|RESEARCH\.md|RELEASE-PLAN\.md" \
     docs/ 2>/dev/null \
     | grep -v "docs/archive/" \
-    | grep -v "specs/archive" | head -20 >&2
+    | grep -v "specs/archive" | sed -n '1,20p' >&2  # sed reads to the end; `head` races pipefail (BUG-2026-11-02-pipefail-pipe-race)
 fi
 
 # ── Epic 1: Skill count consistency ───────────────────────────────────────────
@@ -94,7 +94,7 @@ else
     2>/dev/null \
     | grep -v "specs/archive\|specs/epics/archive\|docs/archive/\|\.git\|node_modules\|\.gemini\|\.cursor" \
     | grep -v "CHANGELOG\.md\|specs/verifications/reports/\|PLAN-evolve-structure\.md" \
-    | head -10 >&2
+    | sed -n '1,10p' >&2
 fi
 
 # ── Epic 3: BCP slop guard ────────────────────────────────────────────────────
@@ -107,7 +107,7 @@ if [[ "$BCP_SLOP" -eq 0 ]]; then
 else
   doctrine_fail "'Build Commit Point' found ($BCP_SLOP hits) — canonical term is 'Business Complexity Points'; see docs/references/bcp.md"
   grep -ri "build commit point" . 2>/dev/null \
-    | grep -v "specs/archive\|\.git\|node_modules\|\.gemini\|\.cursor\|validate-doctrine\.sh" | head -5 >&2
+    | grep -v "specs/archive\|\.git\|node_modules\|\.gemini\|\.cursor\|validate-doctrine\.sh" | sed -n '1,5p' >&2
 fi
 
 # ── Epic 4: SKILL.md size cap ─────────────────────────────────────────────────

package/scripts/validate-skill-catalog.sh

@@ -18,7 +18,7 @@ VIOLATIONS=0
 WARNINGS=0
 
 # Documented naming exceptions (craft-skill REFERENCE.md)
-VERB_NOUN_EXCEPTIONS="grill-me grill-with-docs context7-mcp using-bigpowers deploy generate-allure-report"
+VERB_NOUN_EXCEPTIONS="grill-me grill-with-docs using-bigpowers deploy generate-allure-report"
 
 catalog_usage() {
   echo "Usage: bash scripts/validate-skill-catalog.sh [--archive] [--strict] [--skill <name>]"

package/skills-lock.json

@@ -41,11 +41,6 @@
       "sha256": "5dc23c0642af9d2b",
       "path": "skills/compose-workflow/SKILL.md"
     },
-    "context7-mcp": {
-      "description": "Fetch current library docs via Context7 MCP instead of training data. Use when user asks about frameworks, APIs, setup, or code examples for React, Next.js, Prisma, etc.",
-      "sha256": "f58dcec5aff6a0bf",
-      "path": "skills/context7-mcp/SKILL.md"
-    },
     "craft-skill": {
       "description": "Create new bigpowers skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill for the bigpowers lifecycle.",
       "sha256": "7acb068f458fa1b4",

package/.agents/skills/context7-mcp/SKILL.md

@@ -1,55 +0,0 @@
----
-name: context7-mcp
-model: haiku
-description: "Fetch current library docs via Context7 MCP instead of training data. Use when user asks about frameworks, APIs, setup, or code examples for React, Next.js, Prisma, etc."
----
-
-# Context7 MCP
-
-> **HARD GATE** — Max **3** Context7 tool calls per user question (`resolve-library-id` + `query-docs` count toward the cap). On quota/rate-limit errors, emit an explicit **CONTEXT7_UNAVAILABLE** block — do NOT silently answer from training data.
->
-> **HARD GATE** — Before HTTP fetch, check `bash scripts/lib/doc-fetch-cache.sh get "<libraryId>:<query>"`. Cache hit within TTL → use cached body (no round-trip). On ETag mismatch after conditional refresh, replace cache entry.
-
-## When to Use
-
-- Setup/configuration questions ("How do I configure Next.js middleware?")
-- Code involving libraries ("Write a Prisma query for…")
-- API references ("What are the Supabase auth methods?")
-- User mentions specific frameworks (React, Vue, Svelte, Express, Tailwind, etc.)
-
-## Bounded Retry (max 3x)
-
-| Attempt | Action |
-|---------|--------|
-| 1 | `resolve-library-id` → pick best match |
-| 2 | `query-docs` with selected `libraryId` |
-| 3 | Retry `query-docs` once with refined query (narrower scope) |
-
-After 3 failures, stop and print:
-
-```
-CONTEXT7_UNAVAILABLE
-Reason: <quota|rate-limit|no-match|timeout>
-Action: Ask user to retry later, paste official docs URL, or run `bts docs <lib>`.
-Do NOT substitute training-data answers without labeling them UNVERIFIED.
-```
-
-## Fetch Cache (ETag-revalidated)
-
-1. **Cache key:** `"<libraryId>:<normalized-query>"` (lowercase, trimmed).
-2. **Read:** `bash scripts/lib/doc-fetch-cache.sh get "<key>"` — exit 0 → use cached body.
-3. **Miss / stale:** call `query-docs`; store via `doc-fetch-cache.sh put`.
-4. **TTL:** 300s default (`DOC_CACHE_TTL`). Stale entries refresh on next fetch; honor `ETag` when MCP returns it.
-
-`bts docs <lib>` shares the same cache helper when invoked from this skill.
-
-## Process
-
-1. `resolve-library-id` with `libraryName` + full user `query`.
-2. Select match: name similarity, reputation, benchmark score; prefer version-specific IDs when user names a version.
-3. `query-docs` with `libraryId` + specific question (one concept per call).
-4. Answer using fetched docs; cite library/version when relevant.
-
-## Verify
-
-→ verify: `test -f scripts/lib/doc-fetch-cache.sh`

package/.cline/skills/context7-mcp/SKILL.md

@@ -1,55 +0,0 @@
----
-name: context7-mcp
-model: haiku
-description: "Fetch current library docs via Context7 MCP instead of training data. Use when user asks about frameworks, APIs, setup, or code examples for React, Next.js, Prisma, etc."
----
-
-# Context7 MCP
-
-> **HARD GATE** — Max **3** Context7 tool calls per user question (`resolve-library-id` + `query-docs` count toward the cap). On quota/rate-limit errors, emit an explicit **CONTEXT7_UNAVAILABLE** block — do NOT silently answer from training data.
->
-> **HARD GATE** — Before HTTP fetch, check `bash scripts/lib/doc-fetch-cache.sh get "<libraryId>:<query>"`. Cache hit within TTL → use cached body (no round-trip). On ETag mismatch after conditional refresh, replace cache entry.
-
-## When to Use
-
-- Setup/configuration questions ("How do I configure Next.js middleware?")
-- Code involving libraries ("Write a Prisma query for…")
-- API references ("What are the Supabase auth methods?")
-- User mentions specific frameworks (React, Vue, Svelte, Express, Tailwind, etc.)
-
-## Bounded Retry (max 3x)
-
-| Attempt | Action |
-|---------|--------|
-| 1 | `resolve-library-id` → pick best match |
-| 2 | `query-docs` with selected `libraryId` |
-| 3 | Retry `query-docs` once with refined query (narrower scope) |
-
-After 3 failures, stop and print:
-
-```
-CONTEXT7_UNAVAILABLE
-Reason: <quota|rate-limit|no-match|timeout>
-Action: Ask user to retry later, paste official docs URL, or run `bts docs <lib>`.
-Do NOT substitute training-data answers without labeling them UNVERIFIED.
-```
-
-## Fetch Cache (ETag-revalidated)
-
-1. **Cache key:** `"<libraryId>:<normalized-query>"` (lowercase, trimmed).
-2. **Read:** `bash scripts/lib/doc-fetch-cache.sh get "<key>"` — exit 0 → use cached body.
-3. **Miss / stale:** call `query-docs`; store via `doc-fetch-cache.sh put`.
-4. **TTL:** 300s default (`DOC_CACHE_TTL`). Stale entries refresh on next fetch; honor `ETag` when MCP returns it.
-
-`bts docs <lib>` shares the same cache helper when invoked from this skill.
-
-## Process
-
-1. `resolve-library-id` with `libraryName` + full user `query`.
-2. Select match: name similarity, reputation, benchmark score; prefer version-specific IDs when user names a version.
-3. `query-docs` with `libraryId` + specific question (one concept per call).
-4. Answer using fetched docs; cite library/version when relevant.
-
-## Verify
-
-→ verify: `test -f scripts/lib/doc-fetch-cache.sh`