bigpowers 2.61.4 → 2.63.0

package/.pi/package.json

@@ -1,6 +1,6 @@
 {
   "name": "bigpowers",
-  "version": "2.61.4",
+  "version": "2.63.0",
   "description": "74 skills — 73 agent skills synthesizing 17 years of software engineering discipline into a prescriptive methodology for solo developers",
   "keywords": [
     "pi-package"

package/.pi/prompts/audit-code.md

@@ -68,7 +68,7 @@ Run this self-review before asking anyone else to look at the code. The goal is
 - [ ] Every new function has at least one test
 - [ ] Every bug fix has a regression test
 - [ ] Tests verify behavior through public interfaces (not implementation details)
-- [ ] Tests are F.I.R.S.T compliant (use `enforce-first` if unsure)
+- [ ] Tests are F.I.R.S.T compliant (per CONVENTIONS.md §Tests; use `enforce-first` if unsure)
 
 ### SOLID and Heuristics
 

package/.pi/prompts/enforce-first.md

@@ -1,5 +1,5 @@
 ---
-description: Apply the F.I.R.S.T test quality rubric (Fast, Independent, Repeatable, Self-Validating, Timely) to a test suite or individual tests. Use when develop-tdd is writing tests, when test quality needs to be checked, or when user mentions F.I.R.S.T or "test quality".
+description: Apply the F.I.R.S.T test quality rubric (per CONVENTIONS.md §Tests) to a test suite or individual tests. Use when develop-tdd is writing tests, when test quality needs to be checked, or when user mentions F.I.R.S.T or "test quality".
 ---
 
 
@@ -7,7 +7,7 @@ description: Apply the F.I.R.S.T test quality rubric (Fast, Independent, Repeata
 > **HARD GATE** — **HARD GATE** — Before shipping, ALL enforcement checks must pass: lint, typecheck, tests, coverage gates. Do NOT disable or skip checks to get to green.
 
 
-Apply the F.I.R.S.T rubric (Uncle Bob, Clean Code Chapter 9) to evaluate and improve tests.
+Apply the F.I.R.S.T rubric per CONVENTIONS.md §Tests to evaluate and improve tests.
 
 This skill is typically invoked internally by `develop-tdd` during the test-writing phase. It can also be run standalone on an existing test suite.
 
@@ -18,58 +18,15 @@ This skill is typically invoked internally by `develop-tdd` during the test-writ
 
 ## The F.I.R.S.T Rubric
 
-### F — Fast
+The canonical F.I.R.S.T definition lives in CONVENTIONS.md §Tests. This skill operationalizes each criterion:
 
-Tests must run quickly. Slow tests don't get run. They don't get trusted.
+- **F (Fast):** No real I/O; suite target < 30s
+- **I (Independent):** No shared state; any order, same result
+- **R (Repeatable):** No machine deps; passes identically on CI and locally
+- **S (Self-Validating):** Assertions, not console.log; descriptive failures
+- **T (Timely):** Written with code; regression tests for every fix
 
-- [ ] No real network calls (use fakes/stubs for external I/O)
-- [ ] No real database (use in-memory or transaction-rollback strategies)
-- [ ] No `sleep` or arbitrary timeouts in test code
-- [ ] The full suite runs in under 30 seconds (target; adjust to project size)
-
-**Fix:** Replace slow I/O with named fake classes. Never inline anonymous stubs.
-
-### I — Independent
-
-Tests must not depend on each other. Running in any order must produce the same result.
-
-- [ ] No shared mutable state between tests
-- [ ] Each test sets up its own data and tears it down
-- [ ] No test assumes another test ran first
-- [ ] Tests can be run individually (e.g. `npm test -- mytest.test.ts`) and pass
-
-**Fix:** Move setup into `beforeEach`. Use factory functions to build test data.
-
-### R — Repeatable
-
-Tests must pass consistently in any environment.
-
-- [ ] No dependency on machine-specific paths, ports, or environment variables (unless explicitly injected)
-- [ ] No dependency on current time without mocking the clock
-- [ ] No flakiness — a test that sometimes fails is worse than no test
-- [ ] Tests pass on CI the same way they pass locally
-
-**Fix:** Inject time, randomness, and environment as parameters. Pin seeds for anything random.
-
-### S — Self-Validating
-
-Tests must report pass or fail automatically. No human inspection required.
-
-- [ ] Tests use assertions (`expect`, `assert`, etc.) — not just `console.log`
-- [ ] Failure messages are descriptive enough to diagnose without reading the test body
-- [ ] No tests that "pass" by default when the feature is broken
-
-**Fix:** Add assertion messages. Use matchers that describe the expected behavior.
-
-### T — Timely
-
-Tests must be written at the right time — before or immediately with the code they test.
-
-- [ ] Tests are written in the same commit as the code (or the commit before, if TDD)
-- [ ] No "I'll add tests later" patterns
-- [ ] Bug fixes include a regression test that would have caught the bug
-
-**Fix:** Run `develop-tdd` — it enforces the timely principle by design.
+For the full rubric with checklists and fix patterns, see CONVENTIONS.md §Tests.
 
 ## Applying the rubric
 

package/.pi/prompts/seed-conventions.md

@@ -28,6 +28,8 @@ Ask the user these questions (one at a time, wait for each answer):
 5. **Conventions** — "Any naming, file organization, or patterns all agents must follow?"
 6. **Never-do list** — "What are the hard stops? Things an agent must never touch?"
 7. **Defensive code categories** — "Which apply? (Rate limit / Retry / Circuit breaker / Timeout / Graceful degradation)"
+8. **Local tool wiring (optional)** — "Wire bigpowers skills for project-local tools that global install can't reach? (Cursor + OpenCode only)" If yes, generate the local wiring artifacts described in [REFERENCE.md](REFERENCE.md) §Local tool wiring. If no, skip — the standard seed output is unchanged.
+8. **Local tool wiring (optional)** — "Wire bigpowers skills for project-local tools that global install can't reach? (Cursor + OpenCode only)" If yes, generate the local wiring artifacts described in [REFERENCE.md](REFERENCE.md) §Local tool wiring. If no, skip — the standard seed output is unchanged.
 
 ## Generate files
 
@@ -130,3 +132,34 @@ Use the standard bigpowers CONVENTIONS.md as the base. Fill in the project-speci
 ## Stack profile fragments
 
 If the user selected a stack profile, merge the matching `profiles/<name>.md` fragment into the generated `CONVENTIONS.md` under a `## Stack Conventions` section. Profiles supply language-specific commands, architecture patterns, and never-do additions.
+
+## Local tool wiring (optional interview step 8)
+
+Offered after the standard interview. Covers the two tools that global install (`scripts/install.sh`) structurally cannot reach because they read project-root config, not global paths.
+
+### Cursor — project-local `.cursor/rules` symlink
+
+```bash
+# From the project root:
+ln -sfn <bigpowers-install-path>/.cursor/rules .cursor/rules
+```
+
+Cursor reads `.cursor/rules/` from the project root. This symlink gives every project access to bigpowers skills as Cursor rules without duplicating the files. Run once per project.
+
+### OpenCode — project-local `opencode.json` + `AGENTS.md`
+
+`opencode.json` (project root):
+```json
+{
+  "$schema": "https://opencode.ai/config.json",
+  "instructions": [".cursor/rules/*.mdc", "AGENTS.md"]
+}
+```
+
+OpenCode reads `opencode.json` from the project root, NOT from a global path. The `instructions` array points to the local `.cursor/rules` symlink (from the Cursor step above) and the project's `AGENTS.md`. Both must exist in the project for OpenCode to see bigpowers skills.
+
+`AGENTS.md` is already generated by the standard interview (step 2 of Generate Files). When local tool wiring is opted in, ensure `AGENTS.md` includes the standard agent-config template header `# [Project Name] — OpenCode`.
+
+### When to offer
+
+Only offer local tool wiring when the user's project will be opened in Cursor or OpenCode. These tools are project-root scoped by design — no global installer can solve them. Global install (`install.sh`) already handles Claude Code, Gemini CLI, and pi globally. Do not offer for tools that read global config.

package/.pi/skills/audit-code/SKILL.md

@@ -70,7 +70,7 @@ Run this self-review before asking anyone else to look at the code. The goal is
 - [ ] Every new function has at least one test
 - [ ] Every bug fix has a regression test
 - [ ] Tests verify behavior through public interfaces (not implementation details)
-- [ ] Tests are F.I.R.S.T compliant (use `enforce-first` if unsure)
+- [ ] Tests are F.I.R.S.T compliant (per CONVENTIONS.md §Tests; use `enforce-first` if unsure)
 
 ### SOLID and Heuristics
 

package/.pi/skills/enforce-first/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: enforce-first
-description: "Apply the F.I.R.S.T test quality rubric (Fast, Independent, Repeatable, Self-Validating, Timely) to a test suite or individual tests. Use when develop-tdd is writing tests, when test quality needs to be checked, or when user mentions F.I.R.S.T or \"test quality\"."
+description: "Apply the F.I.R.S.T test quality rubric (per CONVENTIONS.md §Tests) to a test suite or individual tests. Use when develop-tdd is writing tests, when test quality needs to be checked, or when user mentions F.I.R.S.T or \"test quality\"."
 model: haiku
 ---
 
@@ -9,7 +9,7 @@ model: haiku
 > **HARD GATE** — **HARD GATE** — Before shipping, ALL enforcement checks must pass: lint, typecheck, tests, coverage gates. Do NOT disable or skip checks to get to green.
 
 
-Apply the F.I.R.S.T rubric (Uncle Bob, Clean Code Chapter 9) to evaluate and improve tests.
+Apply the F.I.R.S.T rubric per CONVENTIONS.md §Tests to evaluate and improve tests.
 
 This skill is typically invoked internally by `develop-tdd` during the test-writing phase. It can also be run standalone on an existing test suite.
 
@@ -20,58 +20,15 @@ This skill is typically invoked internally by `develop-tdd` during the test-writ
 
 ## The F.I.R.S.T Rubric
 
-### F — Fast
+The canonical F.I.R.S.T definition lives in CONVENTIONS.md §Tests. This skill operationalizes each criterion:
 
-Tests must run quickly. Slow tests don't get run. They don't get trusted.
+- **F (Fast):** No real I/O; suite target < 30s
+- **I (Independent):** No shared state; any order, same result
+- **R (Repeatable):** No machine deps; passes identically on CI and locally
+- **S (Self-Validating):** Assertions, not console.log; descriptive failures
+- **T (Timely):** Written with code; regression tests for every fix
 
-- [ ] No real network calls (use fakes/stubs for external I/O)
-- [ ] No real database (use in-memory or transaction-rollback strategies)
-- [ ] No `sleep` or arbitrary timeouts in test code
-- [ ] The full suite runs in under 30 seconds (target; adjust to project size)
-
-**Fix:** Replace slow I/O with named fake classes. Never inline anonymous stubs.
-
-### I — Independent
-
-Tests must not depend on each other. Running in any order must produce the same result.
-
-- [ ] No shared mutable state between tests
-- [ ] Each test sets up its own data and tears it down
-- [ ] No test assumes another test ran first
-- [ ] Tests can be run individually (e.g. `npm test -- mytest.test.ts`) and pass
-
-**Fix:** Move setup into `beforeEach`. Use factory functions to build test data.
-
-### R — Repeatable
-
-Tests must pass consistently in any environment.
-
-- [ ] No dependency on machine-specific paths, ports, or environment variables (unless explicitly injected)
-- [ ] No dependency on current time without mocking the clock
-- [ ] No flakiness — a test that sometimes fails is worse than no test
-- [ ] Tests pass on CI the same way they pass locally
-
-**Fix:** Inject time, randomness, and environment as parameters. Pin seeds for anything random.
-
-### S — Self-Validating
-
-Tests must report pass or fail automatically. No human inspection required.
-
-- [ ] Tests use assertions (`expect`, `assert`, etc.) — not just `console.log`
-- [ ] Failure messages are descriptive enough to diagnose without reading the test body
-- [ ] No tests that "pass" by default when the feature is broken
-
-**Fix:** Add assertion messages. Use matchers that describe the expected behavior.
-
-### T — Timely
-
-Tests must be written at the right time — before or immediately with the code they test.
-
-- [ ] Tests are written in the same commit as the code (or the commit before, if TDD)
-- [ ] No "I'll add tests later" patterns
-- [ ] Bug fixes include a regression test that would have caught the bug
-
-**Fix:** Run `develop-tdd` — it enforces the timely principle by design.
+For the full rubric with checklists and fix patterns, see CONVENTIONS.md §Tests.
 
 ## Applying the rubric
 

package/.pi/skills/seed-conventions/SKILL.md

@@ -30,6 +30,8 @@ Ask the user these questions (one at a time, wait for each answer):
 5. **Conventions** — "Any naming, file organization, or patterns all agents must follow?"
 6. **Never-do list** — "What are the hard stops? Things an agent must never touch?"
 7. **Defensive code categories** — "Which apply? (Rate limit / Retry / Circuit breaker / Timeout / Graceful degradation)"
+8. **Local tool wiring (optional)** — "Wire bigpowers skills for project-local tools that global install can't reach? (Cursor + OpenCode only)" If yes, generate the local wiring artifacts described in [REFERENCE.md](REFERENCE.md) §Local tool wiring. If no, skip — the standard seed output is unchanged.
+8. **Local tool wiring (optional)** — "Wire bigpowers skills for project-local tools that global install can't reach? (Cursor + OpenCode only)" If yes, generate the local wiring artifacts described in [REFERENCE.md](REFERENCE.md) §Local tool wiring. If no, skip — the standard seed output is unchanged.
 
 ## Generate files
 
@@ -132,3 +134,34 @@ Use the standard bigpowers CONVENTIONS.md as the base. Fill in the project-speci
 ## Stack profile fragments
 
 If the user selected a stack profile, merge the matching `profiles/<name>.md` fragment into the generated `CONVENTIONS.md` under a `## Stack Conventions` section. Profiles supply language-specific commands, architecture patterns, and never-do additions.
+
+## Local tool wiring (optional interview step 8)
+
+Offered after the standard interview. Covers the two tools that global install (`scripts/install.sh`) structurally cannot reach because they read project-root config, not global paths.
+
+### Cursor — project-local `.cursor/rules` symlink
+
+```bash
+# From the project root:
+ln -sfn <bigpowers-install-path>/.cursor/rules .cursor/rules
+```
+
+Cursor reads `.cursor/rules/` from the project root. This symlink gives every project access to bigpowers skills as Cursor rules without duplicating the files. Run once per project.
+
+### OpenCode — project-local `opencode.json` + `AGENTS.md`
+
+`opencode.json` (project root):
+```json
+{
+  "$schema": "https://opencode.ai/config.json",
+  "instructions": [".cursor/rules/*.mdc", "AGENTS.md"]
+}
+```
+
+OpenCode reads `opencode.json` from the project root, NOT from a global path. The `instructions` array points to the local `.cursor/rules` symlink (from the Cursor step above) and the project's `AGENTS.md`. Both must exist in the project for OpenCode to see bigpowers skills.
+
+`AGENTS.md` is already generated by the standard interview (step 2 of Generate Files). When local tool wiring is opted in, ensure `AGENTS.md` includes the standard agent-config template header `# [Project Name] — OpenCode`.
+
+### When to offer
+
+Only offer local tool wiring when the user's project will be opened in Cursor or OpenCode. These tools are project-root scoped by design — no global installer can solve them. Global install (`install.sh`) already handles Claude Code, Gemini CLI, and pi globally. Do not offer for tools that read global config.

package/CHANGELOG.md

@@ -1,3 +1,17 @@
+# [2.63.0](https://github.com/danielvm-git/bigpowers/compare/v2.62.0...v2.63.0) (2026-07-05)
+
+
+### Features
+
+* **e47:** cross-tool skill distribution — pi install, seed wiring, docs, verify harness ([27c68b4](https://github.com/danielvm-git/bigpowers/commit/27c68b49b698e994b783abc85c3750637061664e))
+
+# [2.62.0](https://github.com/danielvm-git/bigpowers/compare/v2.61.4...v2.62.0) (2026-07-05)
+
+
+### Features
+
+* **install:** add pi global install support, remove broken OpenCode step ([d91c86e](https://github.com/danielvm-git/bigpowers/commit/d91c86e46fd85d8fe6a7013484f4dd5ce6bc201f))
+
 ## [2.61.4](https://github.com/danielvm-git/bigpowers/compare/v2.61.3...v2.61.4) (2026-07-04)
 
 

package/SKILL-INDEX.md

@@ -70,7 +70,7 @@
 | 42 | Build | `wire-observability` | Add structured JSON logging, observability commands, and idempotent setup script | ✅ Active |
 | 43 | Verify | `audit-code` | Self-review checklist for the coding agent to run before dispatching a reviewer. | ✅ Active |
 | 44 | Verify | `diagnose-root` | Run 4-phase root cause analysis — reproduce, isolate, hypothesize, verify. Use | ✅ Active |
-| 45 | Verify | `enforce-first` | Apply the F.I.R.S.T test quality rubric (Fast, Independent, Repeatable, Self-Val | ✅ 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 | `inspect-quality` | Interactive QA session where user reports bugs or issues conversationally, and t | ✅ Active |
 | 48 | Verify | `investigate-bug` | Investigate a bug or issue by exploring the codebase to find root cause, then wr | ✅ Active |

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "bigpowers",
-  "version": "2.61.4",
+  "version": "2.63.0",
   "description": "73 agent skills synthesizing 17 years of software engineering discipline into a prescriptive methodology for solo developers",
   "main": "index.js",
   "scripts": {

package/scripts/install.sh

@@ -4,8 +4,8 @@
 # Supported tools:
 #   Claude Code  → ~/.claude/skills/<name>/ (one symlink per skill)
 #   Gemini CLI   → ~/.gemini/extensions/bigpowers/ (one dir symlink)
+#   pi           → ~/.pi/agent/skills/<name>/ (one symlink per skill)
 #   Cursor       → ~/.cursor/rules/ (one dir symlink; per-project note printed)
-#   OpenCode     → instructions printed (no standard global path)
 #
 # Usage:
 #   ./scripts/install.sh              # install
@@ -195,28 +195,33 @@ uninstall_cursor() {
   unlink_if_managed "$CURSOR_RULES_DST" "$REPO_ROOT/"
 }
 
-# ── OpenCode ──────────────────────────────────────────────────────────────────
-
-install_opencode() {
-  local opencode_config="$REPO_ROOT/opencode.json"
-  if [[ ! -f "$opencode_config" ]]; then
-    echo "Creating opencode.json..."
-    {
-      echo "{"
-      echo "  \"\$schema\": \"https://opencode.ai/config.json\","
-      echo "  \"instructions\": [\".cursor/rules/*.mdc\"]"
-      echo "}"
-    } > "$opencode_config"
-  else
-    echo "opencode.json already exists, skipping."
-  fi
+# ── pi ────────────────────────────────────────────────────────────────────────
+
+PI_CONFIG_DIR="$HOME/.pi"
+PI_SKILLS_DIR="$PI_CONFIG_DIR/agent/skills"
+
+install_pi() {
+  echo ""
+  echo "pi → $PI_SKILLS_DIR/"
+  local count=0
+  for skill_dir in "$SKILLS_ROOT"/*/; do
+    [[ -f "$skill_dir/SKILL.md" ]] || continue
+    local name; name="$(basename "$skill_dir")"
+    link "$skill_dir" "$PI_SKILLS_DIR/$name"
+    count=$((count + 1))
+  done
+  echo "  $count skills installed"
 }
 
-print_opencode_instructions() {
+uninstall_pi() {
   echo ""
-  echo "OpenCode — integration active:"
-  echo "  - opencode.json created (points to .cursor/rules/*.mdc)"
-  echo "  - Add AGENTS.md for project-specific rules if needed"
+  echo "pi → removing management from $PI_CONFIG_DIR/"
+  if [[ -d "$PI_SKILLS_DIR" ]]; then
+    for dst in "$PI_SKILLS_DIR"/*/; do
+      [[ -L "${dst%/}" ]] || continue
+      unlink_if_managed "${dst%/}" "$REPO_ROOT/"
+    done
+  fi
 }
 
 # ── main ──────────────────────────────────────────────────────────────────────
@@ -228,15 +233,15 @@ $UNINSTALL && echo "(uninstall mode)"
 if $UNINSTALL; then
   uninstall_claude
   uninstall_gemini
+  uninstall_pi
   uninstall_cursor
   echo ""
   echo "bigpowers uninstalled."
 else
   install_claude
   install_gemini
+  install_pi
   install_cursor
-  install_opencode
-  print_opencode_instructions
   echo ""
   echo "bigpowers installed. Future updates:"
   if [[ -d "$REPO_ROOT/.git" ]]; then

package/scripts/lib/skill-common.sh

@@ -0,0 +1,73 @@
+#!/usr/bin/env bash
+# story: e28s01
+# scripts/lib/skill-common.sh — Shared bash library for skill pipeline scripts
+# Provides: resolve_repo_root, parse_frontmatter, iterate_skills
+# Safe to source under set -euo pipefail; double-source is a no-op.
+
+# Double-source guard
+if [[ -n "${_SKILL_COMMON_SOURCED:-}" ]]; then
+  return 0
+fi
+_SKILL_COMMON_SOURCED=1
+
+# ── resolve_repo_root ───────────────────────────────────────────────
+# Resolves REPO_ROOT to the repository root (two levels up from this
+# library at scripts/lib/skill-common.sh). Also sets SKILLS_ROOT:
+# skills/ subdirectory when it exists, repo root fallback.
+resolve_repo_root() {
+  REPO_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/../.." && pwd)"
+  SKILLS_ROOT="$REPO_ROOT"
+  [[ -d "$REPO_ROOT/skills" ]] && SKILLS_ROOT="$REPO_ROOT/skills"
+}
+
+# Resolve at source time so callers don't need to
+resolve_repo_root
+
+# ── parse_frontmatter ───────────────────────────────────────────────
+# Usage: parse_frontmatter <SKILL.md>
+# Extracts YAML frontmatter into globals: _PF_NAME, _PF_MODEL, _PF_DESCRIPTION
+# Returns 0 on success, non-zero if file missing or no name in frontmatter.
+parse_frontmatter() {
+  local file="$1"
+  _PF_NAME=""
+  _PF_MODEL=""
+  _PF_DESCRIPTION=""
+
+  if [[ ! -f "$file" ]]; then
+    echo "parse_frontmatter: file not found: $file" >&2
+    return 1
+  fi
+
+  # Verify YAML frontmatter exists (at least one --- delimiter)
+  if ! grep -q '^---$' "$file"; then
+    echo "parse_frontmatter: no YAML frontmatter in $file" >&2
+    return 1
+  fi
+
+  # Extract name (single-line, first match in frontmatter)
+  _PF_NAME=$(awk '/^---/{f++} f==1 && /^name:/{print; exit}' "$file" | sed 's/^name:[[:space:]]*//')
+
+  # Extract model (single-line, first match in frontmatter)
+  _PF_MODEL=$(awk '/^---/{f++} f==1 && /^model:/{print; exit}' "$file" | sed 's/^model:[[:space:]]*//')
+
+  # Extract description (potentially multiline, from "description:" to next YAML key or closing ---)
+  _PF_DESCRIPTION=$(awk '/^---/{f++; next} f==1 && /^description:/{p=1; sub(/^description:[[:space:]]*/,""); print; next} f==1 && p && /^[a-z]+:/{exit} f==1 && p{print}' "$file" \
+    | tr -d '\n' \
+    | sed -E 's/[[:space:]]+/ /g')
+
+  [[ -z "$_PF_NAME" ]] && return 1
+  return 0
+}
+
+# ── iterate_skills ──────────────────────────────────────────────────
+# Usage: iterate_skills
+# Outputs one absolute path per skill directory under SKILLS_ROOT that
+# contains a SKILL.md file, in stable sorted order (no trailing slash).
+iterate_skills() {
+  local dir
+  for dir in "$SKILLS_ROOT"/*/; do
+    [[ -d "$dir" ]] || continue
+    [[ -f "$dir/SKILL.md" ]] || continue
+    echo "${dir%/}"
+  done
+}

package/scripts/verify-install.sh

@@ -0,0 +1,60 @@
+#!/usr/bin/env bash
+# story: e47s04
+# verify-install.sh — manual assertion harness for install + seed wiring
+set -euo pipefail
+
+REPO_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+PASS=0
+FAIL=0
+TMPDIR=""
+
+cleanup() { [[ -n "$TMPDIR" && -d "$TMPDIR" ]] && rm -rf "$TMPDIR"; }
+trap cleanup EXIT
+
+pass() { echo "  PASS $1"; PASS=$((PASS + 1)); }
+fail() { echo "  FAIL $1"; FAIL=$((FAIL + 1)); }
+
+echo "=== install.sh --dry-run ==="
+DRY_OUT="$("$REPO_ROOT/scripts/install.sh" --dry-run 2>&1)"
+
+echo "$DRY_OUT" | grep -q 'pi →' && pass "install --dry-run includes pi" || fail "install --dry-run missing pi"
+DRY_UNINSTALL="$("$REPO_ROOT/scripts/install.sh" --dry-run --uninstall 2>&1)"
+echo "$DRY_UNINSTALL" | grep -q 'pi →' && pass "uninstall --dry-run includes pi" || fail "uninstall --dry-run missing pi"
+echo "$DRY_OUT" | grep -qi 'opencode' && fail "install references opencode" || pass "install has no opencode reference"
+echo "$DRY_OUT" | grep -q 'Claude Code →' && pass "install includes Claude Code" || fail "install missing Claude Code"
+echo "$DRY_OUT" | grep -q 'Gemini CLI →' && pass "install includes Gemini CLI" || fail "install missing Gemini CLI"
+echo "$DRY_OUT" | grep -q 'Cursor →' && pass "install includes Cursor" || fail "install missing Cursor"
+
+for tool in "Claude Code" "pi"; do
+  # Find the skills count in the tool's section (between this tool header and next section)
+  section=$(echo "$DRY_OUT" | sed -n "/${tool} →/,/^$/p")
+  count=$(echo "$section" | grep -o '[0-9]\+ skills installed' | grep -o '[0-9]\+' || echo "0")
+  [[ "$count" -gt 0 ]] && pass "${tool}: $count skills" || fail "${tool}: 0 skills"
+done
+
+echo ""
+echo "=== seed-conventions local wiring ==="
+grep -qi 'local tool wiring' "$REPO_ROOT/skills/seed-conventions/SKILL.md" && pass "SKILL.md: local tool wiring" || fail "SKILL.md: missing local tool wiring"
+grep -qi 'opencode.json' "$REPO_ROOT/skills/seed-conventions/REFERENCE.md" && pass "REFERENCE.md: opencode.json" || fail "REFERENCE.md: missing opencode.json"
+grep -qi 'cursor/rules' "$REPO_ROOT/skills/seed-conventions/REFERENCE.md" && pass "REFERENCE.md: Cursor symlink" || fail "REFERENCE.md: missing Cursor symlink"
+
+OPENCODE_JSON=$(sed -n '/```json/,/```/p' "$REPO_ROOT/skills/seed-conventions/REFERENCE.md" | grep -v '```' || true)
+if [[ -n "$OPENCODE_JSON" ]]; then
+  TMPDIR=$(mktemp -d)
+  echo "$OPENCODE_JSON" > "$TMPDIR/test-opencode.json"
+  jq empty "$TMPDIR/test-opencode.json" 2>/dev/null && pass "opencode.json valid JSON" || fail "opencode.json NOT valid JSON"
+else
+  fail "opencode.json template not found"
+fi
+
+echo ""
+echo "=== install.sh source ==="
+grep -q 'install_pi()' "$REPO_ROOT/scripts/install.sh" && pass "source: install_pi()" || fail "source: missing install_pi()"
+grep -q 'uninstall_pi()' "$REPO_ROOT/scripts/install.sh" && pass "source: uninstall_pi()" || fail "source: missing uninstall_pi()"
+grep -q 'PI_SKILLS_DIR="$PI_CONFIG_DIR/agent/skills"' "$REPO_ROOT/scripts/install.sh" && pass "source: targets ~/.pi/agent/skills/" || fail "source: wrong pi target"
+! grep -q 'install_opencode\|print_opencode_instructions' "$REPO_ROOT/scripts/install.sh" && pass "source: no opencode functions" || fail "source: opencode functions remain"
+
+echo ""
+echo "──────────────────────────────────────────"
+echo "verify-install: $PASS passed, $FAIL failed"
+[[ "$FAIL" -gt 0 ]] && { echo "Overall: fail"; exit 1; } || echo "Overall: pass"

package/skills/audit-code/SKILL.md

@@ -70,7 +70,7 @@ Run this self-review before asking anyone else to look at the code. The goal is
 - [ ] Every new function has at least one test
 - [ ] Every bug fix has a regression test
 - [ ] Tests verify behavior through public interfaces (not implementation details)
-- [ ] Tests are F.I.R.S.T compliant (use `enforce-first` if unsure)
+- [ ] Tests are F.I.R.S.T compliant (per CONVENTIONS.md §Tests; use `enforce-first` if unsure)
 
 ### SOLID and Heuristics
 

package/skills/enforce-first/SKILL.md

@@ -2,14 +2,14 @@
 name: enforce-first
 model: haiku
 effort: standard
-description: Apply the F.I.R.S.T test quality rubric (Fast, Independent, Repeatable, Self-Validating, Timely) to a test suite or individual tests. Use when develop-tdd is writing tests, when test quality needs to be checked, or when user mentions F.I.R.S.T or "test quality".
+description: Apply the F.I.R.S.T test quality rubric (per CONVENTIONS.md §Tests) to a test suite or individual tests. Use when develop-tdd is writing tests, when test quality needs to be checked, or when user mentions F.I.R.S.T or "test quality".
 ---
 
 # Enforce FIRST
 > **HARD GATE** — **HARD GATE** — Before shipping, ALL enforcement checks must pass: lint, typecheck, tests, coverage gates. Do NOT disable or skip checks to get to green.
 
 
-Apply the F.I.R.S.T rubric (Uncle Bob, Clean Code Chapter 9) to evaluate and improve tests.
+Apply the F.I.R.S.T rubric per CONVENTIONS.md §Tests to evaluate and improve tests.
 
 This skill is typically invoked internally by `develop-tdd` during the test-writing phase. It can also be run standalone on an existing test suite.
 
@@ -20,58 +20,15 @@ This skill is typically invoked internally by `develop-tdd` during the test-writ
 
 ## The F.I.R.S.T Rubric
 
-### F — Fast
+The canonical F.I.R.S.T definition lives in CONVENTIONS.md §Tests. This skill operationalizes each criterion:
 
-Tests must run quickly. Slow tests don't get run. They don't get trusted.
+- **F (Fast):** No real I/O; suite target < 30s
+- **I (Independent):** No shared state; any order, same result
+- **R (Repeatable):** No machine deps; passes identically on CI and locally
+- **S (Self-Validating):** Assertions, not console.log; descriptive failures
+- **T (Timely):** Written with code; regression tests for every fix
 
-- [ ] No real network calls (use fakes/stubs for external I/O)
-- [ ] No real database (use in-memory or transaction-rollback strategies)
-- [ ] No `sleep` or arbitrary timeouts in test code
-- [ ] The full suite runs in under 30 seconds (target; adjust to project size)
-
-**Fix:** Replace slow I/O with named fake classes. Never inline anonymous stubs.
-
-### I — Independent
-
-Tests must not depend on each other. Running in any order must produce the same result.
-
-- [ ] No shared mutable state between tests
-- [ ] Each test sets up its own data and tears it down
-- [ ] No test assumes another test ran first
-- [ ] Tests can be run individually (e.g. `npm test -- mytest.test.ts`) and pass
-
-**Fix:** Move setup into `beforeEach`. Use factory functions to build test data.
-
-### R — Repeatable
-
-Tests must pass consistently in any environment.
-
-- [ ] No dependency on machine-specific paths, ports, or environment variables (unless explicitly injected)
-- [ ] No dependency on current time without mocking the clock
-- [ ] No flakiness — a test that sometimes fails is worse than no test
-- [ ] Tests pass on CI the same way they pass locally
-
-**Fix:** Inject time, randomness, and environment as parameters. Pin seeds for anything random.
-
-### S — Self-Validating
-
-Tests must report pass or fail automatically. No human inspection required.
-
-- [ ] Tests use assertions (`expect`, `assert`, etc.) — not just `console.log`
-- [ ] Failure messages are descriptive enough to diagnose without reading the test body
-- [ ] No tests that "pass" by default when the feature is broken
-
-**Fix:** Add assertion messages. Use matchers that describe the expected behavior.
-
-### T — Timely
-
-Tests must be written at the right time — before or immediately with the code they test.
-
-- [ ] Tests are written in the same commit as the code (or the commit before, if TDD)
-- [ ] No "I'll add tests later" patterns
-- [ ] Bug fixes include a regression test that would have caught the bug
-
-**Fix:** Run `develop-tdd` — it enforces the timely principle by design.
+For the full rubric with checklists and fix patterns, see CONVENTIONS.md §Tests.
 
 ## Applying the rubric
 

package/skills/seed-conventions/REFERENCE.md

@@ -61,3 +61,34 @@ Use the standard bigpowers CONVENTIONS.md as the base. Fill in the project-speci
 ## Stack profile fragments
 
 If the user selected a stack profile, merge the matching `profiles/<name>.md` fragment into the generated `CONVENTIONS.md` under a `## Stack Conventions` section. Profiles supply language-specific commands, architecture patterns, and never-do additions.
+
+## Local tool wiring (optional interview step 8)
+
+Offered after the standard interview. Covers the two tools that global install (`scripts/install.sh`) structurally cannot reach because they read project-root config, not global paths.
+
+### Cursor — project-local `.cursor/rules` symlink
+
+```bash
+# From the project root:
+ln -sfn <bigpowers-install-path>/.cursor/rules .cursor/rules
+```
+
+Cursor reads `.cursor/rules/` from the project root. This symlink gives every project access to bigpowers skills as Cursor rules without duplicating the files. Run once per project.
+
+### OpenCode — project-local `opencode.json` + `AGENTS.md`
+
+`opencode.json` (project root):
+```json
+{
+  "$schema": "https://opencode.ai/config.json",
+  "instructions": [".cursor/rules/*.mdc", "AGENTS.md"]
+}
+```
+
+OpenCode reads `opencode.json` from the project root, NOT from a global path. The `instructions` array points to the local `.cursor/rules` symlink (from the Cursor step above) and the project's `AGENTS.md`. Both must exist in the project for OpenCode to see bigpowers skills.
+
+`AGENTS.md` is already generated by the standard interview (step 2 of Generate Files). When local tool wiring is opted in, ensure `AGENTS.md` includes the standard agent-config template header `# [Project Name] — OpenCode`.
+
+### When to offer
+
+Only offer local tool wiring when the user's project will be opened in Cursor or OpenCode. These tools are project-root scoped by design — no global installer can solve them. Global install (`install.sh`) already handles Claude Code, Gemini CLI, and pi globally. Do not offer for tools that read global config.

package/skills/seed-conventions/SKILL.md

@@ -30,6 +30,8 @@ Ask the user these questions (one at a time, wait for each answer):
 5. **Conventions** — "Any naming, file organization, or patterns all agents must follow?"
 6. **Never-do list** — "What are the hard stops? Things an agent must never touch?"
 7. **Defensive code categories** — "Which apply? (Rate limit / Retry / Circuit breaker / Timeout / Graceful degradation)"
+8. **Local tool wiring (optional)** — "Wire bigpowers skills for project-local tools that global install can't reach? (Cursor + OpenCode only)" If yes, generate the local wiring artifacts described in [REFERENCE.md](REFERENCE.md) §Local tool wiring. If no, skip — the standard seed output is unchanged.
+8. **Local tool wiring (optional)** — "Wire bigpowers skills for project-local tools that global install can't reach? (Cursor + OpenCode only)" If yes, generate the local wiring artifacts described in [REFERENCE.md](REFERENCE.md) §Local tool wiring. If no, skip — the standard seed output is unchanged.
 
 ## Generate files
 

package/skills-lock.json

@@ -13,7 +13,7 @@
     },
     "audit-code": {
       "description": "Self-review checklist for the coding agent to run before dispatching a reviewer. Checks CONVENTIONS.md compliance, Boy Scout Rule, test coverage, types, and SOLID. Produces a pass/fail checklist. Use before request-review, before committing, or when user asks for a code quality check.",
-      "sha256": "d9f3ed654fdfe3dd",
+      "sha256": "9a4afa72a83882d6",
       "path": "skills/audit-code/SKILL.md"
     },
     "audit-plan": {
@@ -102,8 +102,8 @@
       "path": "skills/elaborate-spec/SKILL.md"
     },
     "enforce-first": {
-      "description": "Apply the F.I.R.S.T test quality rubric (Fast, Independent, Repeatable, Self-Validating, Timely) to a test suite or individual tests. Use when develop-tdd is writing tests, when test quality needs to be checked, or when user mentions F.I.R.S.T or \"test quality\".",
-      "sha256": "f2948df50c0add34",
+      "description": "Apply the F.I.R.S.T test quality rubric (per CONVENTIONS.md §Tests) to a test suite or individual tests. Use when develop-tdd is writing tests, when test quality needs to be checked, or when user mentions F.I.R.S.T or \"test quality\".",
+      "sha256": "20f717d0496ca20f",
       "path": "skills/enforce-first/SKILL.md"
     },
     "evolve-skill": {
@@ -278,7 +278,7 @@
     },
     "seed-conventions": {
       "description": "Generate CLAUDE.md and CONVENTIONS.md for a brand-new project through a brief interview, and create the specs/ directory with evolved bigpowers structure (product/, tech-architecture/, verifications/, epics/archive/). Entry point for greenfield projects. Use when starting a new project from scratch, when user asks to set up AI agent conventions, or when there is no CLAUDE.md yet.",
-      "sha256": "8a0ca3ccd93f42a7",
+      "sha256": "eb4cbca25d54fa81",
       "path": "skills/seed-conventions/SKILL.md"
     },
     "session-state": {