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 ngocsangyem/MeowKit --skill cookgit clone --depth 1 https://github.com/ngocsangyem/MeowKitWrote 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/ngocsangyem/meowkit/cook)<a href="https://agentmods.dev/skills/ngocsangyem/meowkit/cook"><img src="https://agentmods.dev/badge/skills/ngocsangyem/meowkit/cook/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/ngocsangyem/meowkit/cook"><img src="https://agentmods.dev/badge/skills/ngocsangyem/meowkit/cook.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 | $0.04643 |
| Opus 5 | $0.00037 | $0.02322 |
| Sonnet 5 | $0.00015 | $0.00929 |
| Haiku 4.5 | $0.00007 | $0.00464 |
Grade A, and why
mk:cook 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 8d 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 — 285 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cook — Full Implementation Pipeline
Implementation entry adapter. .claude/workflow.yaml is the only lifecycle phase-sequence authority; this skill selects a profile, invokes the relevant owners, and presents the required gates. TDD is opt-in via --tdd.
Usage
/cook <natural language task OR plan path>
/cook "Add user auth" --fast
/cook "Build payment processor" --tdd # Strict TDD enforced
/cook tasks/plans/260329-feature/plan.md --auto
Profiles: quick/--fast (inspect → edit → focused verify → summary), standard (plan if needed → implement → verify → review), release (standard + explicit mk:ship request), and high-assurance (--tdd --strict, with human gates). Flags: --interactive (default) | --fast | --parallel | --auto | --no-test | --tdd | --verify | --strict | --no-strict.
Quick runs do not create a plan, wiki candidate, memory entry, or ship action by default. They still stop if a risk flag or changed contract requires the standard profile.
Modifier flags (layer on any mode): --verify [LIGHT] (light browser/artifact check) | --strict [HEAVY] (full evaluator pass) | --no-strict (suppress auto-strict)
Concrete cost depends on the inner harness, model tier, and target surface; treat [LIGHT] vs [HEAVY] as relative ordering only.
--verify is advisory (does not block ship). --strict is a hard gate (FAIL blocks ship and routes back to Phase 3).
--advice (composable, off by default) — opt-in strategic supervision for one
run. At named checkpoints (GUIDE after Gate 1, RESCUE on a stall or contradiction,
REVIEW after Verify and before the reviewer, RECHECK after a correction; hard cap 5)
the athena agent assesses the situation, recommends an operational path inside the
locked scope, and may return the work for correction. It never approves: Gate 1 and
Gate 2 remain human, the reviewer keeps its verdict, and supervision counts as no
verification. Without the flag there are zero calls and no state is written.
Checkpoints: references/workflow-steps.md → Advice Checkpoints. Contract:
.claude/rules-conditional/advice-supervision-rules.md (load only when the flag is present).
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/failure-catalog.md 2.9 KB
- references/intent-detection.md 6.5 KB
- references/loop-safety-protocol.md 2.7 KB
- references/review-cycle.md 5.2 KB
- references/subagent-patterns.md 9.3 KB
- references/workflow-steps.md 29 KB
- scripts/test-scripts.sh 5.2 KB runs code
- scripts/validate-gate-1.sh 1.4 KB runs code
- scripts/validate-gate-2.sh 2.9 KB runs code
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.
- 8d ago First seen · 285 lines · 75 tokens per session scan A 6900ee03e8d6
mk:cook is a skill published in the GitHub repository ngocsangyem/MeowKit (14 stars, last pushed 1mo ago), licensed MIT. It adds 75 tokens to every session and 4,643 once invoked, about $0.0004 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 2026-09-03.
Other skills, from other repositories
ash-framework
Ash Framework — resources, actions, policies, aggregates, calculations, AshPhoenix.Form, LiveView, migrations. Use when generating resources via mix ash.codegen, editing changes, checks, types, validations, or domain code interfaces.
deploy
Elixir/Phoenix deployment patterns — Dockerfile, fly.toml, runtime.exs, mix release, rel/ overlays. Use when configuring Fly.io, Docker, CI/CD, health checks, or production migrations.
liveview-patterns
Build LiveView: async data (assignasync), PubSub (check connected?), phx-change events, form components/modals/uploads, streams for lists, livepatch. Use when handling interactions, debugging events, or tracking Presence.
tidewave-integration
Tidewave MCP runtime tools — debugging, smoke testing, live state inspection, SQL queries, hex docs. Use when evaluating code in a running Phoenix app.
oban
Oban job processing — workers, perform/1 (OSS) and process/1 (Pro), queues, cron, retries, unique jobs, idempotency, Oban Pro (Workflow, Batch, Chunk, Smart Engine), Testing. Use when writing Oban workers, queue config, or debugging jobs.
perf
Analyze Elixir/Phoenix performance — N+1 queries, assign bloat, ecto optimization, genserver bottlenecks. Use when slowness, timeouts, or high memory reported.