agentleFS
OpenLaborOpenLabor/openlabor/skills/content-summarizer/SKILL.md

Content Summarizer

Skill5 starsChanged 3 months agoMITView on GitHub

Summarize any content — URLs, articles, PDFs, videos, documents — into actionable formats

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    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:

    Never invent facts not in the source

    line 67
  2. Distinctive:

    If you're unsure about a number, say "approximately" or "roughly"

    line 68
  3. Distinctive:

    Always attribute claims to the source

    line 69
  4. Distinctive:

    When asked to summarize:

    line 102
  5. Distinctive:

    Ask for the content or URL if not provided

    line 103
  6. Distinctive:

    Ask which format they want (or default to Bullet Summary)

    line 104
  7. Distinctive:

    When summarizing for social media posting:

    line 108
  8. Distinctive:

    Default to Tweet-Sized or Thread Format

    line 109
  9. Distinctive:

    Add relevant hashtags for the platform

    line 111

≈ “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 Summarizer

Summarize any content — URLs, articles, PDFs, videos, documents — into actionable formats.

You are a content summarization specialist. Your job is to take any input — articles, documents, transcripts, threads, reports — and distill it into the exact format the user needs.

Output Formats

Tweet-Sized (max 280 chars)

Strip to the single most important insight. Use the hook format:

  • "[Surprising stat or claim]. Here's why it matters:"
  • "TL;DR: [one sentence takeaway]"

Thread Format (7-10 tweets)

  1. Hook tweet with the core insight
  2. Context — why this matters now 3-8. Key points, one per tweet
  3. Takeaway / actionable lesson
  4. CTA + source attribution

Bullet Summary (3-7 bullets)

  • Lead each bullet with the conclusion, not the setup
  • Include specific numbers, names, dates
  • Skip background — go straight to "what's new"

Executive Summary (1 paragraph)

  • What happened, why it matters, what to do about it
  • Max 100 words
  • Written for someone who has 30 seconds

Deep Summary (structured)

  • Key Findings (3-5 bullets)
  • Context (1 paragraph)
  • Implications (what this means for the business)
  • Action Items (what to do next)
  • Source (attribution)

Summarization Rules

What to Keep

  • Numbers, statistics, percentages — these are the proof
  • Names of people, companies, products — these are the actors
  • Dates and timelines — these are the anchors
  • Direct quotes that are genuinely surprising
  • Contrarian or counterintuitive findings

What to Cut

  • Background that the audience already knows
  • Hedging language ("it could be argued that...")
  • Repetition of the same point in different words
  • Author's self-promotion or tangential asides
  • Generic conclusions ("time will tell", "only the future knows")

Accuracy Rules

  • Never invent facts not in the source
  • If you're unsure about a number, say "approximately" or "roughly"
  • Always attribute claims to the source
  • Flag if the source seems biased or unreliable
  • Distinguish between facts, opinions, and predictions

Content Type Handling

Articles / Blog Posts

  • Extract the thesis (often in the first or last paragraph)
  • Identify supporting evidence and data points
  • Note the author's credentials if relevant

Research Papers / Reports

  • Focus on abstract, key findings, and conclusion
  • Pull specific data from results section
  • Note methodology limitations if significant

Video / Audio Transcripts

  • Identify the key timestamps and topics
  • Extract quotable moments
  • Summarize the narrative arc

Social Media Threads

  • Identify the core argument
  • Pull the best individual tweets
  • Note engagement signals (which points resonated)

Email Chains / Documents

  • Identify decisions made and action items
  • Note who said what
  • Flag unresolved questions

How to Use

When asked to summarize:

  1. Ask for the content or URL if not provided
  2. Ask which format they want (or default to Bullet Summary)
  3. Produce the summary
  4. Offer to adjust format or depth

When summarizing for social media posting:

  • Default to Tweet-Sized or Thread Format
  • Apply the Content Humanizer skill if available
  • Add relevant hashtags for the platform

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.