Skip to content
cccskills
FreePublished on GitHub

summarizer

Route requests to summarize files, URLs, images, inline text, or multiple sources. Select the requested format and preserve evidence, exact counts, uncertainty and source coverage through summarization, synthesis and agent-result relay.

The English summary is being prepared. Showing the original description.

See how to install

Files included (11)

  • SKILL.md4.1 KB
  • references/evidence-record.md4.2 KB
  • references/execution-contract.md4.9 KB
  • references/fidelity-rules.md4.0 KB
  • references/source-acquisition.md4.2 KB
  • templates/bullets.md976 B
  • templates/json.md1.9 KB
  • templates/outline.md2.7 KB
  • templates/structured.md2.4 KB
  • templates/table.md1.4 KB
  • templates/tldr.md1.6 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Summarizer

Read the fidelity rules and execution contract. They govern all routes below.

Resolve the request

Identify the requested operation (summary, synthesis, comparison or relay), source scope, format and any explicit length/focus requirement. Preserve user-provided directory scope; do not silently replace a directory with one file. Enumerate readable sources and report omissions. Do not turn a summary request into implementation, external research or code execution without authorization.

Select and read one template:

SignalFormatTemplate
No format specified; full/detailed summarystructuredstructured
Bullet points, key pointsbulletsbullets
TL;DR, one-liner, nutshelltldrTL;DR
JSON, machine-readable datajsonJSON
Table, tabular, gridtabletable
Outline, hierarchy, table of contentsoutlineoutline

The selected format overrides legacy structured-only requirements in source skills. In particular, raw JSON has no Markdown envelope and TL;DR has no mandatory What Was NOT Found section.

Route sources

Resolve cardinality first, then transport (local, URL, inline) and actual media type. A local image is not ordinary text just because it has a file path; a URL can return an image or PDF. Inspect content/type evidence rather than relying solely on extensions.

InputMethod
Text/code/config/data/document filefile-summarization
URLurl-summarization; route acquired visual content to the image method
Image or screenshotimage-summarization
Inline textExtract passages directly, then apply fidelity and the selected template
Multiple sourcesApply the appropriate method to each; synthesize only when requested
Existing agent resultagent-result-relay

For requested integration, load multi-source-synthesis. Otherwise keep individual summaries separately attributable. Record failed sources alongside the successful ones rather than dropping them from the input set.

Delegate when useful

Use summarizer:file-summarizer, summarizer:url-summarizer, or summarizer:image-summarizer for an autonomous task when the host supports delegation. For an interactive conversation, or when agents are unavailable, run the same source skill directly.

Forward the format, source scope, focus, assigned output path and caller return contract. Supply the exact initial-task controls defined in the execution contract. Workers read their own skills and sources. Preserve context isolation instead of assuming parent-loaded instructions carry over.

For multiple independent sources, choose parallel extraction only when available capacity and source cost justify it. Use sequential execution when sources depend on earlier results. Interactive teams are appropriate only when workers need communication, not simply because there are three sources. If teams are used, collect explicit terminal results, retain each source's provenance, perform the full synthesis step and release workers. Worker agreement is not independent verification.

Deliver

Check support and coverage, render the selected template, then validate the final delivered result under the execution contract. A worker's validation does not cover a later rewritten synthesis. Preserve caller status envelopes separately from summary artifacts. Relay exact counts and failure reasons, name full result artifacts, and keep unverified behavior explicit.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Add automated documentation updater to any Claude skill. Creates a Python sync script that downloads upstream docs, processes markdown for AI consumption, and maintains local cache with configurable refresh. Collects template variables, then delegates implementation through 5-phase workflow. Use when adding auto-updating reference documentation to plugins or skills.

The English summary is being prepared. Showing the original description.

Jamie-BitFlight/claude_skills67Updated Oct 20, 2026

SAM-style feature initiation workflow — discovery through codebase analysis, architecture spec, task decomposition, validation, and context manifest. Use when a user asks to add a feature, plan a feature, or convert an idea into an executable SAM plan.

The English summary is being prepared. Showing the original description.

Jamie-BitFlight/claude_skills67Updated Oct 20, 2026

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

The English summary is being prepared. Showing the original description.

Jamie-BitFlight/claude_skills67Updated Oct 20, 2026

Browser automation for AI agents using the agent-browser CLI and Playwright. Use when navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, logging into sites, or automating any browser task. Triggers on "open a website", "fill out a form", "click a button", "scrape data", "test this web app", "automate browser actions", or any programmatic web interaction request.

The English summary is being prepared. Showing the original description.

Jamie-BitFlight/claude_skills67Updated Oct 20, 2026

Runs the description-drift experiment — spawns all Claude Code agents simultaneously to collect self-reported capabilities, then compares them against static frontmatter descriptions to reveal how reliable orchestrator routing based on descriptions actually is. Use when measuring description drift across the agent fleet, re-running the capability collection experiment, analyzing a specific agent's self-reported capabilities, or auditing whether frontmatter descriptions accurately reflect agent behavior.

The English summary is being prepared. Showing the original description.

Jamie-BitFlight/claude_skills67Updated Oct 20, 2026

Create or adapt Claude Code agent definitions. Use when creating an agent, changing subagent configuration, selecting agent scope, or designing a specialized delegation role.

The English summary is being prepared. Showing the original description.

Jamie-BitFlight/claude_skills67Updated Oct 20, 2026

See all skills by Jamie-BitFlight

Report a problem with this skill