agentleFS
RylaispiritRylaispirit/rylai-codex-hermes-skills/skills/content-analysis/SKILL.md

content-analysis

Skill56 starsChanged 3 months agoMITView on GitHub

Analyze articles, documents, videos, podcasts, and transcripts for arguments, evidence, themes, assumptions, contradictions, insights, and unanswered questions.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 12 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈12

    Distinctive · estimate

    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:

    Use tools available in the active runtime to fetch, transcribe, or inspect the source.

    line 24
  2. Distinctive:

    Verify that extraction or transcription is usable before analyzing details.

    line 25
  3. Distinctive:

    If access is incomplete, describe the exact boundary of the analysis.

    line 26
  4. Distinctive:

    Test whether evidence actually supports the conclusion.

    line 56
  5. Distinctive:

    Check quotes against the source.

    line 66
  6. Distinctive:

    Check names, numbers, dates, and attribution.

    line 67
  7. Distinctive:

    Remove observations that cannot be traced to the material.

    line 68
  8. Distinctive:

    Keep quotations short and attribute them when the speaker is known.

    line 102
  9. Distinctive:

    Do not invent transcript segments, timestamps, facts, or recommendations.

    line 103
  10. Distinctive:

    Treat promotional claims and personal anecdotes as evidence of limited type.

    line 105
  11. Distinctive:

    When a source cannot be opened, request the text or transcript instead of presenting a complete analysis.

    line 107
  12. Distinctive:

    Every quoted or quantitative detail is traceable.

    line 114

≈ “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

Content Analysis

Turn source material into a structured explanation of what it says, how it supports its claims, what follows from it, and where uncertainty remains.

Runtime Compatibility

  • Resolve local files from the current workspace or this skill directory.
  • Use tools available in the active runtime to fetch, transcribe, or inspect the source.
  • Verify that extraction or transcription is usable before analyzing details.
  • If access is incomplete, describe the exact boundary of the analysis.

Select The Analysis Lens

Choose only the lenses that serve the request:

  • Argument: claims, reasons, evidence, and conclusion.
  • Editorial: structure, clarity, pacing, framing, and audience fit.
  • Research: method, evidence quality, limits, and competing explanations.
  • Practical: decisions, habits, procedures, and applications.
  • Comparative: agreement, conflict, and differences in assumptions.
  • Media: speaker roles, sequence, emphasis, and transcript reliability.

Workflow

  1. Identify the source
    • Record title, creator, date, format, and available sections when known.
    • Distinguish original material from commentary, reposts, and excerpts.
  2. Check completeness
    • For video or audio, confirm transcript coverage and speaker labels.
    • For documents, check page or section coverage.
    • For web material, note paywalls, dynamic content, or missing attachments.
  3. Map the content
    • State the central question or purpose.
    • Break the material into major claims, themes, or stages.
    • Attach supporting evidence to the claim it is meant to support.
  4. Evaluate
    • Test whether evidence actually supports the conclusion.
    • Identify assumptions, omitted alternatives, contradictions, and ambiguity.
    • Separate source statements from your own inference.
  5. Extract value
    • Rank the strongest insights.
    • Convert applicable ideas into concrete actions or questions.
    • Preserve important limits and conditions.
  6. Verify the report
    • Check quotes against the source.
    • Check names, numbers, dates, and attribution.
    • Remove observations that cannot be traced to the material.

Default Output

## Overview
[Purpose and central message]

## Main Ideas
1. [Claim or theme]
2. [Claim or theme]

## Evidence And Reasoning
- [Claim]: [support and evaluation]

## Strongest Insights
- [Source-grounded insight]

## Assumptions Or Gaps
- [Missing evidence, ambiguity, or alternative explanation]

## Practical Use
- [Action, decision, or question]

## Source Limits
- [Access, extraction, transcription, or scope limitation]

Change this structure when the user asks for a critique, comparison, timeline, study guide, action plan, or another specific deliverable.

Source Integrity

  • Quote only words present in the inspected source.
  • Keep quotations short and attribute them when the speaker is known.
  • Do not invent transcript segments, timestamps, facts, or recommendations.
  • Label deductions as analysis rather than source claims.
  • Treat promotional claims and personal anecdotes as evidence of limited type.
  • For multiple sources, identify which source supports each disputed point.
  • When a source cannot be opened, request the text or transcript instead of presenting a complete analysis.

Final Check

  • The report answers the user's actual analytical question.
  • Summary, evidence, inference, and opinion remain distinguishable.
  • Every quoted or quantitative detail is traceable.
  • Important counterevidence and limitations are visible.
  • Action items follow from the analysis rather than generic advice.

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.