agentleFS
Sign inStart free
harshil-1411harshil-1411/claude_plugin_video_studio/skills/verify/SKILL.md

verify

Skill0 starsChanged 47 days agoApache-2.0View on GitHub

Verify that a video-studio spec is grounded in its sources - which ContentIR claims and evidence each scene cites, which key claims no scene covers, which scenes make statements without claim_refs, and unresolved refs - and fix what it finds by editing the spec. Use when the user runs /video-studio:verify, asks whether a video's claims are backed by the sources, or before rendering a video with grounding strict.

188 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

7 rules. 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 167 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Verify checks project/video-spec.json against the project's ContentIR (source/content-ir.json): every scene's claim_refs must resolve to an …

    line 11
  2. Very specific:

    Use the project folder the user named, else the cwd (absolute path).

    line 18
  3. Distinctive:

    Run this loop when verify returns errors, or warnings the user wants cleared:

    line 28
  4. Very specific:

    Read qa/verify.md for the per-scene and per-claim tables.

    line 30
  5. Distinctive:

    If no span supports it, reword the voiceover or on-screen text to what the sources do say.

    line 37
  6. Distinctive:

    Verify again. Stop when there are no errors, after the third pass, or when a finding repeats unchanged after its fix: then show the …

    line 47
  7. Very specific:

    Do not change grounding to make findings go away unless the user asks for it.

    line 51

“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 claim coverage

Verify checks project/video-spec.json against the project's ContentIR (source/content-ir.json): every scene's claim_refs must resolve to an evidence ref or claim id, and scenes that state things must cite the evidence for them. grounding in the spec sets how strict it is: strict makes an uncited scene an error, loose a warning, off only lists it. CTA and end card scenes need no refs; an uncited hook is always only a warning.

  1. Use the project folder the user named, else the cwd (absolute path). It needs project/video-spec.json, and source/content-ir.json from ingest (without it nothing can be checked; under strict that is an error).
  2. Call mcp__plugin_video-studio_engine__verify {project_dir}.
  3. Report the status (pass, warn, fail), the coverage line (claims covered, evidence cited), then each finding as severity id scene/claim: message, followed by its fix.

Fix loop (at most 3 passes)

Run this loop when verify returns errors, or warnings the user wants cleared:

  1. Read qa/verify.md for the per-scene and per-claim tables. Apply each finding's fix to project/video-spec.json, editing only the fields it names and keeping every other value:
    • ungrounded_scene: add the evidence ref(s) that support the scene's text to its claim_refs (the fix suggests the nearest ones; confirm the span says what the scene says with source_section {project_dir, id: <the ref>}, which returns the section containing it). If no span supports it, reword the voiceover or on-screen text to what the sources do say. Never add a ref that does not support the text.
    • semantic on claim_refs.N: replace the ref with the one the fix names (the nearest existing ref), or remove it.
    • semantic quantitative claim: cite the evidence for the number, or remove the number from the text.
    • uncovered_key_claim: cite the claim in the scene that says it, or ask the user whether to add a scene or drop the key message from the brief.
    • no_content_ir / invalid_content_ir: ask the user to run ingest on the sources (the ingest skill); do not invent evidence refs.
  2. Verify again. Stop when there are no errors, after the third pass, or when a finding repeats unchanged after its fix: then show the remaining findings and ask the user how to proceed.

Do not change grounding to make findings go away unless the user asks for it. Machine-readable results are in qa/verify.json.

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.

Check your setup, free