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 randommonicle/claude-skills --skill safe-smokesgit clone --depth 1 https://github.com/randommonicle/claude-skillsWrote 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/randommonicle/claude-skills/safe-smokes)<a href="https://agentmods.dev/skills/randommonicle/claude-skills/safe-smokes"><img src="https://agentmods.dev/badge/skills/randommonicle/claude-skills/safe-smokes/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/randommonicle/claude-skills/safe-smokes"><img src="https://agentmods.dev/badge/skills/randommonicle/claude-skills/safe-smokes.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 | $0.00100 | $0.01392 |
| Opus 5 | $0.00050 | $0.00696 |
| Sonnet 5 | $0.00020 | $0.00278 |
| Haiku 4.5 | $0.00010 | $0.00139 |
Grade A, and why
safe-smokes scanned grade A 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 11d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Use `localhost`, not `127.0.0.1`. Do not background a curl-poll loop to wait for the dev server; grep the server's own ready signal in its log instead. How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Safe smokes against shared data
A smoke must leave shared state exactly as it found it. On 2026-05-18, before this discipline was written, a smoke deleted ASH's real provider row. The rules below exist so that never recurs.
When this applies
Any test, smoke, or seed that connects to a shared or live database. In PropOS this is the Supabase smoke-db project running against real firm data. The risk is highest for the regulated c1-* smokes and anything touching firm, provider, ledger, demand, or transaction rows.
Confirm the target before you run
More than one project exists per environment, and the active credentials or the Dashboard can point at the wrong one. Before running any smoke that can write, confirm which project the run targets and that it is the one you intend, against the project ref recorded in the repo's own instructions rather than one memorised here. The 2026-05-18 incident combined a destructive delete with an unconfirmed target: flip-and-restore removes the first risk, and confirming the target removes the second.
Core rules
- Never
.delete()shared real-firm state. Capture the current value, mutate it, assert, then restore. The test owns only the rows it created. - Flip-and-restore in
try/finally. The restore runs infinallyso it happens even when the assertion throws. A restore that only runs on the happy path is not a restore. - Seed in
beforeAll, tear down inafterAll. Set-up creates the test's own rows; tear-down removes exactly those and nothing else. - NULL foreign keys before deleting parents. For example, NULL
invoices.transaction_idbefore deleting transactions, and widen the capture to the direct FK (payment_authorisations.transaction_id) so nothing is orphaned or blocked. - Union-of-anchors tear-down. Delete by the union of every anchor a test could have created, not just the happy-path id. A test that branches creates different rows on different paths.
- Idempotent and self-cleaning. Do not assume a clean database. Clean up after yourself; never truncate or reset shared tables to force a known state.
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.
- 11d ago First seen · 74 lines · 100 tokens per session scan A bff3fdaaef28
safe-smokes is a skill published in the GitHub repository randommonicle/claude-skills (23 stars, last pushed 4d ago), licensed Apache-2.0. It adds 100 tokens to every session and 1,392 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
review-prs
Review a GitHub pull request in the googleapis/mcp-toolbox repo against the team's reviewer checklist: PR title/description conventions, linked issue, logic errors and unhandled edge cases, breaking changes, test coverage, docs updates, security (input handling), and new dependencies. Use whenever a maintainer asks…
fix-failing-tests
Diagnose a failing test in the googleapis/mcp-toolbox repo and land a fix by reasoning from the actual error: read the failure, reproduce it, shrink it until the cause is forced into the open, then fix the cause. Use this whenever a test or CI job is red, a build breaks after a change, many packages fail at once, or a…
azure-cosmos-db-py
Build Azure Cosmos DB NoSQL services with Python/FastAPI following production-grade patterns. Use when implementing database client setup with dual auth (DefaultAzureCredential + emulator), service layer classes with CRUD operations, partition key strategies, parameterized queries, or TDD patterns for Cosmos. Triggers…
adding-dbt-unit-test
Creates unit test YAML definitions that mock upstream model inputs and validate expected outputs. Use when adding unit tests for a dbt model or practicing test-driven development (TDD) in dbt.
gh-issue
Size-audit, write, and split BanyanDB issues that somebody else or an automated TDD workflow can implement. Use whenever the user asks to file or revise an issue, decide whether an issue is too large, make an issue TDD-ready, turn a design into tickets, or split an umbrella into executable leaves. Do not draft or file…
Drizzle ORM Testing
Testing patterns for Drizzle ORM covering migration testing, query builder testing, transaction testing, and database integration testing with PostgreSQL, SQLite, and MySQL.