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/paultyng/skill-issue/active-readnpx skills add paultyng/skill-issue --skill active-readgit clone --depth 1 https://github.com/paultyng/skill-issueWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/paultyng/skill-issue/active-read)<a href="https://agentmods.dev/skills/paultyng/skill-issue/active-read"><img src="https://agentmods.dev/badge/skills/paultyng/skill-issue/active-read.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.00091 | $0.02320 |
| Opus 5 | $0.00046 | $0.01160 |
| Sonnet 5 | $0.00018 | $0.00464 |
| Haiku 4.5 | $0.00009 | $0.00232 |
Grade A, and why
active-read 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 — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Active Read
Drive an interactive study session over a document or topic via active recall: ingest → section-by-section summary → typology-driven understanding questions → user-led pace → synthesis. Honest evaluation, not flattery.
Workflow
1. Intake
Capture four dimensions before ingesting the source:
- source — file path, URL, Notion page, or pasted prose.
- goal — decide / understand / evaluate / brief someone else.
- time budget — 15 min / 1 hour / multi-session.
- familiarity — cold / warm / refresher (warm = some prior exposure; refresher = solid prior exposure, want a recap).
Source detection rules:
- Path-looking arg with a file ext (e.g.
.pdf,.md,.txt) → local file; read via theReadtool. - URL schema (
http://,https://) → web source; fetch viaWebFetch. notion.so/host or Notion URL pattern → Notion page; fetch viamcp__notion__notion-fetch.- Else → pasted prose fallback; treat the user's text in-conversation as the source.
If the user invokes with no source, ask once and stop until they provide one. Do not infer.
2. Ingest & map
Read the source and produce a section map before any questions.
Section detection:
- First-pass: split on
##headings (level-2 markdown). Each heading + its body is one section. - Fallback when no
##headings exist: split into ~2k-char chunks at paragraph boundaries.
Chunking for large sources:
- For sources >15k chars, fan out parallel Haiku subagents on ~6k-char chunks. Each subagent produces a per-section "what + why this matters" one-line summary. Merge results into a single section map.
- Smaller sources: read inline and produce the map directly.
Thesis extraction (one-shot, here):
Extract the source's central thesis — the one sentence the author would write on a sticky note if asked "what's the point?" — and store it in conversation state for use in synthesis (Task 6 / Section 4). Do NOT re-extract at synthesis time. Anchoring risk: re-extracting after the user has written their own summary risks the agent anchoring on the user's phrasing rather than the source's actual claim. Extract once, freeze.
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.
- 3d ago First seen · 138 lines · 0 tokens per session scan A 44cbffd3bfde
active-read is a skill published in the GitHub repository paultyng/skill-issue (9 stars, last pushed 22d ago), licensed MIT. It adds 91 tokens to every session and 2,320 once invoked, about $0.0005 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
integrate
Map reading notes to thesis chapters and integrate key arguments into the manuscript. Use after completing a reading session to weave new material into the thesis.
academic-cv
Write academic CVs, cover letters, research statements, teaching philosophies, diversity statements, and prepare for academic job applications. Use this skill whenever you need to create or improve an academic CV, write a research statement, develop a teaching philosophy, prepare application materials for faculty…
svm
Skill "svm" from helius-labs/core-ai, covering svm — understand solana's architecture, mcp router surface, prerequisites, how to answer a question and routing.
libby-book-monitor
Track book availability on Libby/OverDrive libraries. Search library catalogues, manage a watchlist, and get notified when books are added. Use for "libby", "check libby", "libby watchlist", "is book on libby", "book available", "overdrive", "library catalogue", "ספרייה", "ספרים".
yt-quiz
YouTube URL을 입력받아 자막을 파싱하고 핵심 내용 요약 + 개념 이해 4지선다 퀴즈를 생성한다. "/yt-quiz", "유튜브 퀴즈", "YouTube 요약" 요청에 사용.
quiz
/dotfiles/memory/questions/ 디렉토리에서 퀴즈를 출제합니다.