Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/mitsuhiko/agent-stuff/summarizenpx skills add mitsuhiko/agent-stuff --skill summarizegit clone --depth 1 https://github.com/mitsuhiko/agent-stuffWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00035 | $0.00549 |
| Opus 5 | $0.00017 | $0.00275 |
| Sonnet 5 | $0.00007 | $0.00110 |
| Haiku 4.5 | $0.00003 | $0.00055 |
Grade A, and why
summarize scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Turn “things” (URLs, PDFs, Word docs, PowerPoints, HTML pages, text files, etc.) into Markdown so they can be inspected/quoted/processed like normal text.
markitdown can fetch URLs by itself; this skill mainly wraps it to make saving + summarizing convenient.
For PDF inputs, use the markitdown[pdf] extra (or the wrapper below, which now does this automatically).
When to use
Use this skill when you need to:
- pull down a web page as a document-like Markdown representation
- convert binary docs (PDF/DOCX/PPTX) into Markdown for analysis
- quickly produce a short summary of a long document before deeper work
Quick usage
Convert a URL or file to Markdown
Run from this skill folder (the agent should cd here first):
uvx --from 'markitdown[pdf]' markitdown <url-or-path>
To write Markdown to a temp file (prints the path) use the wrapper:
node to-markdown.mjs <url-or-path> --tmp
Tip: when summarizing, the script will always write the full converted Markdown to a temp .md file and will always print a final "Hint" line with the path (so you can open/inspect the full content).
Write Markdown to a specific file:
uvx --from 'markitdown[pdf]' markitdown <url-or-path> > /tmp/doc.md
Convert + summarize with haiku-4-5 (pass context!)
Summaries are only useful when you provide what you want extracted and the audience/purpose.
node to-markdown.mjs <url-or-path> --summary --prompt "Summarize focusing on X, for audience Y. Extract Z."
Or:
node to-markdown.mjs <url-or-path> --summary --prompt "Focus on security implications and action items."
This will:
- convert to Markdown via
uvx --from 'markitdown[pdf]' markitdown - write the full Markdown to a temp
.mdfile and print its path as a "Hint" line - run
pi --model claude-haiku-4-5(no-tools, no-session) to summarize using your extra prompt
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 60 lines · 35 tokens per session scan A 3100cd6a8b45
summarize is a skill published in the GitHub repository mitsuhiko/agent-stuff (3,032 stars, last pushed yesterday), licensed Apache-2.0. It adds 35 tokens to every session and 549 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.