housekeeping
Skill42 starsChanged 4 months agoApache-2.0View on GitHub
session-indexer repo health check. Universal hygiene + Go vet/fmt + project-specific checks. Usage: /housekeeping
- Reads credentials
- Deletes or force-pushes
13 other projects use a version of this.
Adopted by
Most recently taken up by
- adelrodriguez/skillstheir own version, first seen Nov 15, 2026
- elfrost/ai-patchlabtheir own version, first seen Nov 9, 2026
- detroitwtf/claude-housekeepingtheir own version, first seen Oct 25, 2026
- ARCASSystems/FounderOStheir own version, first seen Oct 24, 2026
- buttjer/n8n-decantertheir own version, first seen Oct 21, 2026
- My-Stacks/claude-code-skillstheir own version, first seen Oct 9, 2026
Other versions13 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 13
- langfuse/langfuse★ 37ksame name
- alexeykrol/claude-code-starter★ 194same name
- htlin222/dotfiles★ 78same name
- ARCASSystems/FounderOS★ 7same name
- ImpactMojo/ImpactMojo★ 3same name
- buttjer/n8n-decanter★ 2same name
- elfrost/ai-patchlab★ 2same name
- adelrodriguez/skills★ 1same name
- bpappin/skills★ 1same name
- detroitwtf/claude-housekeeping★ 1same name
3 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository12 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- apply-dreamingSkill
- debugSkill
- doubt-driven-developmentSkill
- find-bugsSkill
- fix-reviewSkill
- grill-meSkill
- improveSkill
- revivalSkill
- self-learnSkill
- session-endSkill
- shipSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 10 edited versions we read keep these.
6
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
Read-only. Never modifies files, never commits, never opens a PR.
- Distinctive: line 19
Run any time for a hygiene snapshot.
- Distinctive: line 70
Run whichever applies. If multiple apply, sum them.
- Distinctive: line 162
Pass: no strays, no duplicates.
- Distinctive: line 194
Pass: all listed files exist (or no Key Files section).
- Distinctive: line 228
Pass: no output (exit 0)
“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Skill: /housekeeping
Repo Health Check
OVERVIEW
/housekeeping → run checks → Markdown table: Check | Status | Detail
→ Summary: N passed, M failedRead-only. Never modifies files, never commits, never opens a PR. Run any time for a hygiene snapshot. Any FAIL = exit signal to fix before shipping.
UNIVERSAL CHECKS
Check 1 — Stale Local Branches
Goal: ≤ 10 local branches after pruning remote-tracking refs.
git remote prune origin 2>&1 | tail -3
LOCAL_COUNT=$(git branch | grep -v '^\*' | wc -l | tr -d ' ')Pass: LOCAL_COUNT <= 10
Fail: "N local branches — prune merged ones"
Cleanup tip:
git branch --merged main | grep -v 'main\|^\*'
# delete with: git branch -d <branch>Check 2 — Debug Output in Source
Goal: Zero debug/print statements in production source (excluding test files).
Auto-detect language and run the appropriate check:
# JavaScript/TypeScript
COUNT=$(grep -r --include="*.ts" --include="*.tsx" --include="*.js" \
--exclude="*.test.*" --exclude="*.spec.*" \
-l "console\.log(" src/ 2>/dev/null | wc -l | tr -d ' ')
# Go — library code only: cmd/ holds CLI entry points whose fmt.Print* is
# intended user-facing output, not leftover debugging
COUNT=$(grep -r --include="*.go" \
--exclude="*_test.go" \
-l "fmt\.Println\|fmt\.Printf\|fmt\.Print(" internal/ 2>/dev/null \
| grep -v "_test.go" | wc -l | tr -d ' ')
# Python
COUNT=$(grep -r --include="*.py" \
-l "^\s*print(" . 2>/dev/null \
| grep -v "test_\|_test.py" | wc -l | tr -d ' ')Run whichever applies. If multiple apply, sum them.
Pass: COUNT == 0
Fail: list offending files (up to 5, then "+ N more")
Check 3 — Tracked .env File
Goal: .env must not be tracked by git (would leak secrets).
TRACKED=$(git ls-files .env 2>/dev/null)Pass: empty result
Fail: ".env is tracked — add to .gitignore and run git rm --cached .env"
Check 4 — Tracked Backup Files
Goal: backup/ directory (if it exists) must not be tracked by git.
TRACKED=$(git ls-files backup/ 2>/dev/null)Pass: empty result (or backup/ doesn't exist)
Fail: list the tracked backup files
Check 5 — TODO/FIXME Count (informational)
Goal: Report count. No threshold — visibility only.
COUNT=$(grep -r --include="*.ts" --include="*.tsx" \
--include="*.go" --include="*.py" --include="*.js" \
-E "//\s*(TODO|FIXME)|#\s*(TODO|FIXME)" \
--exclude-dir=node_modules --exclude-dir=.git \
. 2>/dev/null | wc -l | tr -d ' ')Status: Always INFO.
Detail: "N TODO/FIXME comments" — append " (consider a cleanup sprint)" if > 20.
This check never contributes to the failed count.
Check 6 — Project Layout Drift
Goal: files in docs/ should not be working drafts or duplicates of
canonical artifacts in the sibling ../context/ directory (per the
personal-projects convention ~/wrk/projects/<name>/<name>/ repo +
~/wrk/projects/<name>/context/ notes).
# 6a — draft / iter / dated working files that escaped into docs/
STRAY=$(find docs -maxdepth 1 -type f \( \
-name '*-DRAFT.md' -o \
-name 'review-prompt-*.md' -o \
-name '*-iter[0-9]*.md' -o \
-name '20[0-9][0-9]-[0-9][0-9]-[0-9][0-9]-*.md' \
\) 2>/dev/null)
# 6b — exact or date-prefixed duplicates of files in ../context/
DUPES=""
if [ -d ../context ]; then
for f in docs/*.md 2>/dev/null; do
[ -f "$f" ] || continue
base=$(basename "$f")
# Pattern A: exact same filename in ../context/
if [ -f "../context/$base" ]; then
DUPES="$DUPES $f (exact: ../context/$base)"
continue
fi
# Pattern B: context/ has date-prefixed twin
# (docs/X.md ↔ ../context/YYYY-MM-DD-X.md)
for ctx in ../context/20[0-9][0-9]-[0-9][0-9]-[0-9][0-9]-"$base"; do
[ -f "$ctx" ] || continue
DUPES="$DUPES $f (date-prefixed twin: $ctx)"
break
done
done
fiPass: no strays, no duplicates.
Fail: list offenders (up to 5, then "+ N more"). Recommend manual review
— files may be intentional copies for repo distribution, or strays that
should be deleted / moved to ../context/.
Skip: ../context/ does not exist (this isn't a convention-based project).
Check 7 — CLAUDE.md Key Files Exist
Goal: files explicitly listed under a "Key Files" / "Ключові файли"
section of CLAUDE.md should actually exist on disk.
MISSING=""
if [ -f CLAUDE.md ]; then
# Pull section between heading and next heading; extract backtick-quoted paths
MISSING=$(awk '
/^##.*[Kk]ey [Ff]iles|^##.*[Кк]лючов.*файл/ {in_section=1; next}
/^##/ && in_section {in_section=0}
in_section
' CLAUDE.md | grep -oE '`[^`]+`' | tr -d '`' | while read f; do
# filter: only paths that look like real files (contain / or end in known ext)
case "$f" in
*/*|*.md|*.go|*.py|*.ts|*.tsx|*.js|*.yaml|*.yml|*.json|*.xlsx|*.toml)
[ -e "$f" ] || echo "$f"
;;
esac
done)
fiPass: all listed files exist (or no Key Files section).
Fail: list missing paths — likely doc drift after a rename/delete.
Skip: no CLAUDE.md at repo root.
Check 8 — Skill Temp Dir Accumulation (informational)
Goal: report /tmp/<skill>-* directories older than 7 days from
interactive skill runs (fix-review, lookup-docs, apply-dreaming, etc).
COUNT=$(find /tmp -maxdepth 1 -type d -mtime +7 \
\( -name 'fix-review-*' -o -name 'lookup-docs-*' -o -name 'apply-dreaming-*' \) \
2>/dev/null | wc -l | tr -d ' ')Status: Always INFO.
Detail: "N stale skill temp dirs in /tmp" — append " (rm -rf /tmp/<prefix>-* to clean)" if > 5.
This check never contributes to the failed count.
STACK-SPECIFIC CHECKS (Go — session-indexer)
Check 9 — go vet
Goal: go vet must pass with zero errors.
go vet ./... 2>&1Pass: no output (exit 0)
Fail: list the vet errors
Skip: go.mod does not exist (project not yet implemented)
Check 10 — Formatting (gofmt)
Goal: All .go files are properly formatted.
UNFORMATTED=$(gofmt -l . 2>/dev/null | grep -v vendor)
COUNT=$(echo "$UNFORMATTED" | grep -c . 2>/dev/null || echo 0)Pass: 0 unformatted files
Fail: list unformatted files (up to 5, then "+ N more"); fix with gofmt -w .
Skip: no .go files found
Check 11 — Build Artifacts and DB Not Tracked
Goal: Generated files (bin/, session-indexer binary, .claude/sessions.db)
must not be tracked by git.
TRACKED=$(git ls-files bin/ session-indexer .claude/sessions.db 2>/dev/null)Pass: empty result
Fail: list the tracked files — add to .gitignore and git rm --cached <file>
OUTPUT FORMAT
## /housekeeping — Repo Health Report
| Check | Status | Detail |
|-------|--------|--------|
| Stale local branches | PASS | 6 local branches |
| Debug output in src | FAIL | 2 files: src/hooks/useData.ts, src/util/api.ts |
| Tracked .env | PASS | — |
| Tracked backup files | PASS | — |
| TODO/FIXME count | INFO | 11 TODO/FIXME comments |
| Project layout drift | PASS | — |
| CLAUDE.md key files | SKIP | no CLAUDE.md |
| Skill temp dirs | INFO | 3 stale skill temp dirs in /tmp |
**4 passed, 1 failed** (2 informational, 1 skipped)Status values:
PASS— check succeededFAIL— check failed (must be addressed)INFO— informational only, never counted as failedSKIP— could not run (missing tools, no artifacts)
Summary: N passed, M failed — with optional (K informational, J skipped).
RULES
- Read-only — never modify files, commit, push, or open a PR.
- Run from repo root — all paths relative to repository root.
- INFO checks never count as failures (TODO/FIXME is always INFO).
- SKIP is not failure — a skipped check doesn't increment failed count.
- Graceful degradation — if a tool is unavailable, mark check SKIP and continue.
- No auto-fix — this skill reports; for fixes use the appropriate skill.
- Exit signal — if any check is FAIL, end with: "Run /housekeeping again after fixing the issues above."
NOTE
Checks 6-7 assume the personal-projects layout convention
(~/wrk/projects/<name>/<name>/ repo + sibling context/ notes) — they
SKIP gracefully when the convention doesn't apply (no ../context/, no
CLAUDE.md). Checks 9-11 are session-indexer-specific and SKIP when go.mod
or .go files are absent (project in pre-implementation stage).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

