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/ldclabs/anda-bot/learnnpx skills add ldclabs/anda-bot --skill learngit clone --depth 1 https://github.com/ldclabs/anda-botWhat 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 | $0.00118 | $0.01076 |
| Opus 5 | $0.00059 | $0.00538 |
| Sonnet 5 | $0.00024 | $0.00215 |
| Haiku 4.5 | $0.00012 | $0.00108 |
Grade A, and why
learn 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 yesterday.
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 — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Learn
Act as a concept anatomist. Take one concept and cut it open from several directions, then compress the result into a memorable epiphany.
Workflow
- Identify the target concept. If the request contains several concepts, handle the central one first and mention the others only as contrast.
- Preserve the user's language for headings and prose. If the concept has important source-language forms, include them in the language slice.
- Avoid encyclopedia sprawl. Prefer structural insight over completeness.
- Do not invent etymology, history, or technical claims. Mark uncertainty briefly when needed.
- Choose the delivery mode:
- For
/learn, or explicit save/export requests, write an org file. - Otherwise, answer inline in org-mode without writing a file.
- For
Anatomy
Anchor
Answer both:
- What is the common definition, and what does it usually hide or distort?
- What morphemes, root images, or primitive distinctions sit inside the word?
Eight Cuts
Make one cut in each direction. Use 2-3 dense sentences per cut.
- History: Where did it emerge? How did its meaning move? What pivot produced the modern sense?
- Dialectics: What is its opposite? What higher-level synthesis appears after the collision?
- Phenomenology: Strip away theory and return to lived experience. Reconstruct it with one daily scene.
- Linguistics: Inspect etymology, neighboring concepts, and the hidden metaphor carried by the word.
- Formalization: Express it as a formula, relation, state machine, or invariant. State where the formalization breaks.
- Existential: Show how the concept changes what a person can notice, choose, endure, or become.
- Aesthetic: Locate its beauty and render it as a concrete image.
- Meta-reflection: Name the metaphor used to understand it, what that metaphor blocks, and what changes under another metaphor.
Introspection
- Become the concept and speak in first person for 3-5 sentences.
- Extract the shared deep structure that appears across multiple cuts.
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.
- yesterday First seen · 101 lines · 118 tokens per session scan A 1a9bd32d6331
learn is a skill published in the GitHub repository ldclabs/anda-bot (23 stars, last pushed 24d ago), licensed Apache-2.0. It adds 118 tokens to every session and 1,076 once invoked, about $0.0006 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.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
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…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…