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/redhuntlabs/wizard/summarizing-a-documentnpx skills add redhuntlabs/wizard --skill summarizing-a-documentgit clone --depth 1 https://github.com/redhuntlabs/wizardWhat 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 | $0.00029 | $0.00629 |
| Opus 5 | $0.00015 | $0.00315 |
| Sonnet 5 | $0.00006 | $0.00126 |
| Haiku 4.5 | $0.00003 | $0.00063 |
Grade A, and why
summarizing-a-document 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Summarizing a Document
What this does
Produces a short, scannable summary of a longer source. The summary's central claims are each backed by a verbatim quote so the reader can verify quickly.
When to use
- Sharing a long article, paper, or transcript with someone who won't read it
- Capturing what you read so you can refer back without re-reading
- Preparing for a meeting where the source will come up
What you bring (Inputs)
- The source (article, paper, transcript, doc)
- The intended reader (so the level matches)
- A length budget (default: 5 bullets)
What you get (Output)
A markdown summary with: a one-sentence takeaway at the top, then 3-5 bullets, each pairing a claim (in your words) with a verbatim quote from the source.
How it works (Steps)
- Read the full source once.
- Identify the 3-5 most important claims (not the most interesting prose).
- For each claim, find one verbatim quote that supports it. Copy-paste, do not retype.
- Write the bullet: claim in your own words, then the quote in italics with location (page, paragraph, or timestamp).
- Write the one-sentence takeaway at the top: what would the reader do differently after reading this summary?
Quality bar
- Within length budget
- Every claim has a verbatim quote (no paraphrasing posing as quote)
- Quotes are accurate (copy-paste, not retyped from memory)
- Takeaway is action-oriented, not descriptive
Variations
- Pure summary (no quotes): drop step 3-4's quote portion; useful for very informal sharing
- Disagreement summary: add a final bullet "what's contested in this source"
- Comparison summary: when summarizing 2+ sources, end with a one-line comparison
Example
Input: A 2,000-word article on remote work productivity for an engineering manager.
Output:
Takeaway: Async-first teams should default to written status updates over meetings; the data shows 25% productivity gains but only when managers also reduce 1:1 cadence.
- Async beats sync for individual contributor productivity — "Engineers in async-first teams shipped 25% more features per sprint than peers in sync-first teams." (para 4)
- The gain disappears if managers don't also adjust — "Managers who maintained weekly 1:1s erased the async productivity gain." (para 7)
- ...
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.
- 3d ago First seen · 67 lines · 29 tokens per session scan A 077c485b216f
summarizing-a-document is a skill published in the GitHub repository redhuntlabs/wizard (9 stars, last pushed 4mo ago), licensed MIT. It adds 29 tokens to every session and 629 once invoked, about $0.0001 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 2026-08-31.
Other skills, from other repositories
paper-checking
论文查重系统 - 一亿字次级论文库秒级查重，支持纵向查重和横向查重.
ai-provider-claude-vision
Image understanding and document analysis with Claude's multimodal capabilities -- image input formats, PDF processing, multi-image patterns, structured extraction, and token cost estimation.
cre-document-ingestion
CRE Document Ingestion suite — 4 specialist skills for classifying and extracting structured data from deal documents including rent rolls, T-12 financials, and offering memoranda.
requirements-analysis-plus
Use this skill when you need to parse Word/HTML/JSON/Markdown/Excel requirements and produce a structured analysis; triggers include requirements analysis plus and requirement document parsing.
github-figure-bed
Upload images to any GitHub repo as a figure bed and get CDN/markdown links (jsdelivr/jsdmirror/raw). One-time setup.sh guides gh login, auto-detects the owner, creates the default repo (img.shenzjd.com) if missing, and syncs config with the img.shenzjd.com web app via the repo's .imgx-config/config.json — upload with…
fastapi
FastAPI - 高性能 Python Web API 框架.