Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add T4LEL/Claude-Arsenal --skill maintaingit clone --depth 1 https://github.com/T4LEL/Claude-ArsenalWrote 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/t4lel/claude-arsenal/maintain)<a href="https://agentmods.dev/skills/t4lel/claude-arsenal/maintain"><img src="https://agentmods.dev/badge/skills/t4lel/claude-arsenal/maintain/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/t4lel/claude-arsenal/maintain"><img src="https://agentmods.dev/badge/skills/t4lel/claude-arsenal/maintain.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 | $0.00039 | $0.00753 |
| Opus 5 | $0.00019 | $0.00377 |
| Sonnet 5 | $0.00008 | $0.00151 |
| Haiku 4.5 | $0.00004 | $0.00075 |
Grade A, and why
maintain scanned grade A with 0 findings 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 9d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Maintain
Keep a live product healthy with a repeatable monthly sweep — don't wait for it to break.
Copy this checklist and check off items as you complete them:
Maintain Progress:
- [ ] Step 1: Health snapshot
- [ ] Step 2: Dependencies
- [ ] Step 3: Verify after updates (mandatory)
- [ ] Step 4: Data safety
- [ ] Step 5: Sweep
- [ ] Report
Step 1 — Health snapshot
Pull platform logs for fresh errors (vercel logs, supabase MCP get_logs). Note the current error rate and compare hosting/API costs (Vercel, Supabase, any metered LLM usage) against last period — real numbers or stop and say the dashboard wasn't checked. Any figure not pulled from a live source this run is an unverified assumption; label it as such in the report.
Step 2 — Dependencies
Run security advisories first — npm audit / pip-audit. Critical vulnerabilities get fixed now, no exceptions. Apply patch and minor updates in the same pass. List major updates as candidates only — never auto-apply one. Exception to the ordering: when a critical vulnerability's only fix is a major bump, that upgrade task starts now instead of waiting as a candidate. Delegate to the refactor-engineer agent with the current version, target version, and changelog/migration notes (fetched via the context7 MCP tools) as its own task; a major upgrade needs its own plan and test pass, not a slot in a routine sweep.
Step 3 — Verify after updates (mandatory)
Run tests, typecheck, and a production build after any dependency change. Paste the real output — not "should be fine." A failure rolls the update back immediately; debug on a branch rather than leaving a broken bump sitting on top of a working product.
Step 4 — Data safety
Confirm backups exist, then restore-test one thing on a branch or staging — never against production. Supabase: use PITR or a branch restore. An untested backup is a hope, not a backup; if the restore-test fails, or hasn't been run this cycle, say so plainly instead of assuming it would work.
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.
- 9d ago First seen · 52 lines · 39 tokens per session scan A 165615da487d
maintain is a skill published in the GitHub repository T4LEL/Claude-Arsenal (1 stars, last pushed 1mo ago), licensed MIT. It adds 39 tokens to every session and 753 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
workers-best-practices
Cloudflare Workers best practices for production applications. Use when writing, reviewing, or configuring Workers.
find-journalists
Build, refine, dedupe, and enrich small fit-checked journalist lists for newsjack campaigns. Uses the newsjack CLI (preferred) or the medialyst MCP for news search and journalist enrichment, and falls back to a best-effort local mode with no verified contacts; the agent owns how returned data is organized.
story-origin-check
Recover the first public timestamp and canonical major coverage for a newsjacking signal, then decide whether newer coverage is the same story, a different story, or a materially new development.
relevance-coarse-filter
Cheap, high-recall first-pass filter that removes obvious junk from a detector candidate pool before expensive story-origin research and PR judgment. Decides keep, monitoronly, or reject — never ranks, writes angles, verifies dates, or decides whether to pitch.
annotating-task-lineage
Annotate Airflow tasks with data lineage using inlets and outlets. Use when the user wants to add lineage metadata to tasks, specify input/output datasets, or enable lineage tracking for operators without built-in OpenLineage extraction.
checking-freshness
Quick data freshness check. Use when the user asks if data is up to date, when a table was last updated, if data is stale, or needs to verify data currency before using it.