agentleFS
Galbaz1Galbaz1/video-research-mcp/skills/gr-advisor/SKILL.md

gr-advisor

Skill23 starsChanged 39 days agoMITView on GitHub

Recommends an available research workflow when the user asks about Gemini-powered research, YouTube video analysis, web content extraction, or Weaviate knowledge queries. Activates only when the request matches this plugin and no specific workflow was already chosen — not for code editing, debugging, testing, git operations, or general questions.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Search the web for current information

    line 17
  2. Distinctive:

    Do NOT activate for: code tasks, git operations, file editing, general questions, unrelated work, or when the user already selected a …

    line 20
  3. Very specific:

    Check prior work: call knowledge_search(query="<user topic>", limit=3) — skip if Weaviate is unavailable or the query is a system task

    line 25
  4. Distinctive:

    Recommend: use structured format below, max 3 options

    line 26
  5. Distinctive:

    Wait: do not execute — let the user confirm or adjust

    line 27
  6. Distinctive:

    Keep the recommendation bounded to three options and wait for the user's choice.

    line 55

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

GR Workflow Advisor

Recommend an available workflow before executing research, video analysis, or content tasks.

When to Activate

Activate when the user expresses intent to:

  • Research a topic, question, or concept
  • Analyze a YouTube video or local video file
  • Analyze a URL, document, or pasted content
  • Search the web for current information
  • Find or manage past research/knowledge

Do NOT activate for: code tasks, git operations, file editing, general questions, unrelated work, or when the user already selected a workflow. Never recommend the advisor itself — if intent is unclear, ask a clarifying question.

Workflow

  1. Check available routes and categorize intent: research | video | content | knowledge | system. In Codex, discover the installed plugin's MCP tools and recommend their actual exposed names. Use /gr:* routes only when the client registers those commands; Codex does not load the package's Claude commands.
  2. Check prior work: call knowledge_search(query="<user topic>", limit=3) — skip if Weaviate is unavailable or the query is a system task
  3. Recommend: use structured format below, max 3 options
  4. Wait: do not execute — let the user confirm or adjust

Recommendation Format

RECOMMENDED: <available tool or registered command> — <arguments>
WHY: <one sentence>
ALTERNATIVE: <available alternative>
COST: check provider pricing | TIME: depends on scope
NEXT STEP: <follow-up action>

Key Routing Rules

Choose only a route present in the current client. If none is available, report that missing capability instead of inventing a command or starting installation.

IntentCodex MCP toolRegistered Claude command
Quick current questionweb_search/gr:search
Video URLvideo_analyze or video_create_session/gr:video or /gr:video-chat
Document/URLcontent_analyze or research_document/gr:analyze or /gr:research-doc
Topic researchresearch_deep or research_web/gr:research or /gr:research-deep
Prior work existsknowledge_search/gr:recall
Persist completed researchknowledge_ingest/gr:ingest

A quick question does not need long-running deep research. Provider billing can apply to search, analysis and research; explain the selected scope before execution. Keep the recommendation bounded to three options and wait for the user's choice.

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.