agentleFS
tandryukhatandryukha/aidemo/.claude/skills/verify/SKILL.md

verify

Skill32 starsChanged 7 days agoMITView on GitHub

Verify an engine change end-to-end by rendering the bundled fixture (examples/local-demo) and inspecting the output MP4 and intermediates. Use when the user says "verify", "run the smoke test", "e2e this change", "does the fixture still render", or before committing a nontrivial engine change. Knows the gotchas: OPENAI_API_KEY (or a local OPENAI_BASE_URL server) is needed for voice/captions but not probe/record/compose, stage-by-stage re-runs beat full re-renders, AIDEMO_KEEP_TMP=1 keeps .compose-tmp for compose debugging, logs/fail-*.png from failed takes. Do NOT use for: recording real product demos (use record-demo) or doc-only changes with no runtime surface.

  • Reads credentials

189 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use instinct-export, instinct-import and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • TheKREATORSSkill

    instinct-export

    Export instincts from project/global scope to a file

    Also in 3 projects that use this

  • TheKREATORSSkill

    instinct-import

    Import instincts from file or URL into project/global scope

    Also in 3 projects that use this

  • AviadPSkill

    risk-assess

    Assess the risk and blast radius of a proposed code change before implementing it. Evaluates CI impact, framework-wide vs localized scope, rollback path, and identifies blindspots. Use when user says "what's the risk", "is this safe", "assess the impact", "risk assess", "blast radius", or before applying framework-wide or CI-affecting changes. Do NOT use for code quality review (use code-quality) or plan validation (use design-validator).

    Also in 3 projects that use this

  • diego-tobalinaSkill

    build-fix

    Fix build and compilation errors systematically. Run when build fails.

    Also in 6 projects that use this

  • AaravChadhaSkill

    secure

    Security review with a confidence gate - a finding exists only with a concrete exploit scenario and a high/medium/low confidence rating; everything else goes to a worth-hardening list, never inflated. Sweeps five surfaces: auth gates, secrets hygiene, injection and unsafe sinks, unsafe deserialization/crypto/transport, and LLM tool-use trust boundaries. Reports only, never fixes. Use when the user asks for a security review or to check vulnerabilities, secrets, or auth.

    Also in 3 projects that use this

  • ArvorCoSkill

    specify

    Create feature specification from natural language description. Use when starting a new feature or creating a spec. Triggers on: create spec, specify feature, new feature spec.

    Also in 10 projects that use this

  • +1 more from TheKREATORS

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 167 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If neither is available, prefer the offline full e2e — the bundled mock serves both endpoints locally, so every stage still runs, with zero …

    line 43
  2. Distinctive:

    Don't leave the fixture server running or the port collides on the next run.

    line 81

“Common” is measured: kept by at least 6 in 10 of the 167 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

verify — e2e smoke test for engine changes

Your job: prove a change to the engine still renders the self-contained fixture end-to-end. Be terse; surface only failures + a final status.

Phases

  1. Typecheck (always):
    npm run typecheck
  2. Fixture server (background; it's a plain static server on :8787):
    node examples/local-demo/serve.mjs &
    curl -sf http://localhost:8787 >/dev/null && echo up
  3. Full e2e (needs OPENAI_API_KEY in .env, or OPENAI_BASE_URL pointing at a local OpenAI-compatible server — check with node bin/aidemo.mjs doctor first; it reports the resolved endpoint):
    node bin/aidemo.mjs render examples/local-demo --headless
    If neither is available, prefer the offline full e2e — the bundled mock serves both endpoints locally, so every stage still runs, with zero API spend (say which path you used in the report):
    node test/mock-openai.mjs &            # mock TTS+STT on :8790
    OPENAI_API_KEY= OPENAI_BASE_URL=http://127.0.0.1:8790/v1 \
      node bin/aidemo.mjs render examples/local-demo --headless
    (Narration is sine tones and captions are placeholder words — fine for verifying the pipeline, useless for judging output quality.) Last resort: probe (selectors) + record + compose against previously generated audio//generated/ artifacts if present (captions --offline can regenerate caption cues from an existing voice.json without network).
  4. Inspect the output — existence is not enough:
    ffprobe -v error -show_entries format=duration -show_entries stream=codec_type \
      -of default=nw=1 examples/local-demo/output/final-demo.mp4
    Expect one video + one audio stream and a plausible duration (the fixture is ~30–60 s). A suspiciously tiny file or missing audio stream is a failure.
  5. Stage-scoped iteration — if the change touches one stage, re-run just it (voice, record, captions, compose, or probe for selector work) instead of a full re-render; each stage is independently re-runnable.
  6. Cleanup: kill the serve process from phase 2.

Gotchas

  • Logs: examples/local-demo/logs/<command>.log; failed takes leave logs/fail-*.png/json — read those before re-running blind.
  • Compose hangs → look for an unbounded apad or -stream_loop -1 in the filtergraph (they never EOF; see AGENTS.md ffmpeg gotchas).
  • Captions missing from the output with no error → negative-pts stream-copy concat; the card assembly concat must re-encode.
  • AIDEMO_KEEP_TMP=1 node bin/aidemo.mjs compose examples/local-demo keeps .compose-tmp/ so you can ffprobe intermediate segments.
  • Don't leave the fixture server running or the port collides on the next run.

Report

verify: PASS|FAIL
typecheck: ok|errors
e2e: full render | offline full render (mock) | keyless fallback (probe+record+compose) | skipped (why)
output: <duration>s, video+audio | anomaly: <what>
failures: <stage → cause → log excerpt>   (omit if none)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.