Borrowing it
Nothing to install: this file belongs to akopichin/afm. 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/akopichin/afm/main/.claude/skills/verify/SKILL.mdgit clone --depth 1 https://github.com/akopichin/afmWrote 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/akopichin/afm/verify)<a href="https://agentmods.dev/skills/akopichin/afm/verify"><img src="https://agentmods.dev/badge/skills/akopichin/afm/verify/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/akopichin/afm/verify"><img src="https://agentmods.dev/badge/skills/akopichin/afm/verify.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 | $0.00024 | $0.01308 |
| Opus 5.5 | $0.00010 | $0.00523 |
| Sonnet 5.5 | $0.00005 | $0.00262 |
| Haiku 4.5 | $0.00002 | $0.00131 |
Grade C, and why
verify scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
Docker mode mounts `$(pwd)` into the container 1:1. If you `rm -rf .afm/runs` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
dashboard through the browser or curl. This exercises the real Go How it starts
The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verifying afm end-to-end
Build → run a flow.yaml with a mock agent (no real API calls) → drive the
dashboard through the browser or curl. This exercises the real Go
orchestrator + pkg/server + dashboard, not just go test.
Build
make docker-build # tags akopichin/afm:latest from Dockerfile.runtime, local single-arch
Sanity check: docker run --rm akopichin/afm:latest --version (note: the
entrypoint IS afm, so pass flags directly — afm afm --version is wrong).
Mock agent instead of real claude
Real claude needs CLAUDE_CODE_OAUTH_TOKEN/ANTHROPIC_API_KEY inside
Docker (macOS Keychain isn't reachable from the Linux container) — skip
this entirely for orchestrator-only smoke tests with a bash mock agent that
speaks claude's stream-json protocol:
jq -nc --arg t "some text" '{type:"assistant",message:{content:[{type:"text",text:$t}]}}'
Phase is inferred from context afm itself provides, not from prompt wording:
- implementation: the prompt always embeds a literal
Stage directory for .done file: <path>line (pkg/orchestrator/agents.gorunImplementationAgent) → write.donethere. - autonomous (
agents: [auto]):AFM_STAGE_DIRenv var is set for this phase and the prompt mentionsexecution_summary.md→ write it there. - interactive dialog (any phase with
AFM_STAGE_DIRset, i.e.interactive: truestages): write<phase>.<id>.question.jsoninto$AFM_STAGE_DIR, poll for<phase>.<id>.answer.json, then proceed. - default (non-interactive planning/review): just emit assistant text
with
## Tasks/## Assumptions/## Acceptance Criteriaheadings —RunPlanningwrites the collected text toplan.mditself when noWritetool_use event matches;RunAgent(review) ignores the text entirely.
One unified script covers every stage type — see the phase-inference order above (check implementation's marker first, then autonomous's, then interactive, then default).
Project-level .afm/config.yaml for a smoke-test project dir
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.
- 2mo ago First seen · 112 lines · 24 tokens per session scan C 412193f7218b
verify is a skill published in the GitHub repository akopichin/afm (25 stars, last pushed today), licensed MIT. It adds 24 tokens to every session and 1,308 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…
test-master
Generates test files, creates mocking strategies, analyzes code coverage, designs test architectures, and produces test plans and defect reports across functional, performance, and security testing disciplines. Use when writing unit tests, integration tests, or E2E tests; creating test strategies or automation…
eval-dataset-design
A guide to creating reliable test sets for AI agents. Each evaluation task includes the request, the simulated user's behaviour, a reset starting state, and a success check that can be independently verified.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
test-warp-ui
Guides testing Warp UI features and changes using the computer use tool. Use this skill only when computer-use testing was requested (explicit request or accepted offer) and the computeruse tool is available to the agent. Covers launching Warp and verifying UI behavior.