Borrowing it
Nothing to install: this file belongs to syndicalt/zaxy. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/syndicalt/zaxy/master/.claude/skills/store-doctor/SKILL.mdgit clone --depth 1 https://github.com/syndicalt/zaxyWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/syndicalt/zaxy/store-doctor)<a href="https://agentmods.dev/skills/syndicalt/zaxy/store-doctor"><img src="https://agentmods.dev/badge/skills/syndicalt/zaxy/store-doctor/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/syndicalt/zaxy/store-doctor"><img src="https://agentmods.dev/badge/skills/syndicalt/zaxy/store-doctor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00075 | $0.01361 |
| Opus 5 | $0.00037 | $0.00681 |
| Sonnet 5 | $0.00015 | $0.00272 |
| Haiku 4.5 | $0.00007 | $0.00136 |
Grade B, and why
store-doctor scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
grep -A4 '"Stop"' .claude/settings.local.json How it starts
The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Store doctor — Zaxy memory incident response
Prime directive: .eventloom/*.jsonl is the hash-chained source of truth —
never edit or delete it. Everything under .eventloom/projections/ is a
rebuildable projection: the fix is almost always move aside + rebuild, and
"move aside" is reversible, so it never needs permission-level caution.
Second directive: never open a suspect store in your main process. A corrupt
store can hang ~2 minutes and then segfault whatever opened it. All probes
below run in timeout-wrapped subprocesses.
1. Triage — gather without touching
cd <workspace>
grep -E '^EVENTLOOM_PATH|^PROJECTION_BACKEND' .env 2>/dev/null
ls -la .eventloom/ && ls -la .eventloom/projections/
pgrep -af "zaxy serve"
For each serve PID, establish ownership before judging it:
tr '\0' '\n' < /proc/<PID>/environ | grep EVENTLOOM_PATH
Only PIDs bound to THIS store are suspects. Never kill serve processes belonging to other workspaces.
2. Read the signature
| Observation | Diagnosis |
|---|---|
Multiple serve PIDs, same EVENTLOOM_PATH |
Orphaned multi-owner (writers corrupt the WAL) |
Tiny embedded.kuzu (KBs) + large .kuzu.wal (100KB+) |
Orphan-WAL corruption; open throws RuntimeError ... wal_record.cpp ... UNREACHABLE_CODE |
Huge embedded.kuzu (100s of MB) vs small JSONL logs + tiny clean WAL |
Bloat variant — open hangs ~2min then segfaults; bypasses the self-heal (native crash, no Python exception) |
| Checkout "hangs" / MCP disconnects repeatedly | Either of the above; do steps 3-4 before theorizing |
Checkout works but returns only hook.stop/reminder noise |
Not corruption — capture gap; go to step 7 |
Sanity anchor: a healthy projection for ~500KB of JSONL is ~tens of KB.
3. Rule out the library (30s, subprocess)
timeout 30 python - <<'PY'
import tempfile, os, faulthandler; faulthandler.enable()
import ladybug
d = tempfile.mkdtemp()
db = ladybug.Database(os.path.join(d, "fresh.kuzu"))
conn = ladybug.Connection(db)
conn.execute("CREATE NODE TABLE T(id INT64, PRIMARY KEY(id))")
print("FRESH DB OK — ladybug native is healthy; the problem is the store file")
PY
Fresh DB fails too ⇒ broken native install (reinstall zaxy-memory /
ladybug), stop here. Fresh DB fine ⇒ the store is the patient.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 126 lines · 75 tokens per session scan B 3e97496705f0
store-doctor is a skill published in the GitHub repository syndicalt/zaxy (17 stars, last pushed 1mo ago), licensed MIT. It adds 75 tokens to every session and 1,361 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
connection-troubleshooting
Use this skill when diagnosing PostgreSQL connection issues through PGAutoPilot. Trigger when user asks to: - "I can't connect to the database" - "Connection refused" - "Connection timed out" - "Database is not reachable" - "How do I fix my database URL?" - "The database won't start" - Connection errors…
query-optimization
Use this skill when analyzing, optimizing, or debugging PostgreSQL query performance. Trigger when user asks to: - "Why is this query slow?" - "Help me optimize this query" - "What indexes would improve performance?" - "Analyze the execution plan" - "Find slow queries" - "Tune database performance" - "My app is slow"…
triage-issues
Triage GitHub issues in the googleapis/mcp-toolbox repo: propose the correct labels (type / priority / product / status), check for duplicates, verify a bug has enough info to act on, and draft a triage comment. Use whenever a maintainer asks you to triage, label, categorize, prioritize, or "look at" an issue (or a…
postgresql-indexing
PostgreSQL indexing best practices for Prowler: index design, partial indexes, partitioned table indexing, EXPLAIN ANALYZE validation, concurrent operations, monitoring, and maintenance. Trigger: When creating or modifying PostgreSQL indexes, analyzing query performance with EXPLAIN, debugging slow queries, reviewing…
graphjin-eval
Create, extend, run, baseline, and diagnose GraphJin agent evaluations through the graphjin eval CLI.
ix
This skill should be used when answering structural questions about a codebase: understanding what a symbol is, tracing flows, measuring change impact, finding callers/callees/imports, or detecting code smells. It drives the Ix CLI (ix map/explain/trace/impact/search/rank/smells) against a persistent code graph stored…