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 skills add craigcossairt/trellis --skill learngit clone --depth 1 https://github.com/craigcossairt/trellisWrote 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/craigcossairt/trellis/learn)<a href="https://agentmods.dev/skills/craigcossairt/trellis/learn"><img src="https://agentmods.dev/badge/skills/craigcossairt/trellis/learn.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 5.1 | $0.00058 | $0.00952 |
| Opus 5 | $0.00029 | $0.00476 |
| Sonnet 5 | $0.00012 | $0.00190 |
| Haiku 4.5 | $0.00006 | $0.00095 |
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 6d 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are closing the learning loop. Review this conversation for things worth capturing, then update the right artifacts so the learning survives into future sessions.
What to look for
Scan the conversation and classify any of these as candidates to capture:
- Bug patterns - a symptom someone ran into, a root cause, and a fix that should be visible to the next person who sees the same symptom.
- Tool gotchas - something about a tool, framework, or service that surprised you and would surprise the next session.
- Conventions - a decision about how code should look or how work should flow in this repo, validated in this session.
- Cross-session knowledge - user preferences, working-style feedback, project milestones.
- Context drift - anything you noticed is stale in existing docs.
Things to SKIP:
- Transient task state ("we're in the middle of X")
- Stuff already documented elsewhere (check before duplicating)
- Vague observations ("this could be cleaner")
- Conversation turns that were noise
Where each type goes
| Type | Target | Action |
|---|---|---|
| Bug pattern | docs/common-gotchas.md |
Append a row using the file's format. Include commit SHA + issue ID if known. Auto-apply. |
| Tool gotcha | docs/common-gotchas.md (or agent memory if not project-specific) |
Auto-apply. |
| Convention | AGENTS.md |
Propose the diff to the user first - do NOT auto-edit AGENTS.md. |
| Cross-session knowledge | Your harness's persistent memory, if available | Auto-apply per its conventions. |
| Context drift | Flag to the user | Don't fix silently; say what's stale and where. |
Process
- Read the conversation above. Identify 0-5 capture candidates. Don't force it - if nothing's worth saving, say so and exit.
- Supersession pass (write-time invalidation). Before writing each capture, search for what
it touches: grep
common-gotchas.mdand memory for the same symptom/topic (and query the project brain, if initialized). Three outcomes:- Already documented and still true - skip, or fold new detail into the existing entry.
- Documented but now contradicted or outdated - update the OLD artifact in the same
session. Correct it in place, or when the old fact has historical value, mark it superseded
instead: if your memory format supports metadata, add
superseded_by: <successor>and keep the file; for gotchas/docs, edit in place - git history preserves the old text. Never write the new fact and leave the contradicted one live; retrieval and future greps will keep serving it. - Net-new - write fresh.
- For auto-apply categories (bug patterns, tool gotchas, cross-session knowledge): make the edits, then list them in the output.
- For propose-first categories (conventions): show the proposed diff and ask for approval before editing.
- At the end, output a short summary:
- Captured: X entries applied (list files + one-line descriptions)
- Superseded: entries invalidated/updated by this session's captures (list old -> new)
- Proposed: Y edits waiting on approval
- Drift flagged: Z (list files that look stale)
- Nothing worth capturing: if that was the outcome, say so plainly.
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.
- 6d ago First seen · 73 lines · 58 tokens per session scan A 5f7667190c8b
learn is a skill published in the GitHub repository craigcossairt/trellis (4 stars, last pushed 17d ago), licensed MIT. It adds 58 tokens to every session and 952 once invoked, about $0.0003 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
lov-auto-context
Manual or hook-triggered context operator. Evaluates the current session for pollution (long conversations, topic drift, stale noise), AND takes concrete context-shaping actions: writing project memory, updating global or project CLAUDE.md (with diff + confirm), and recommending harness commands like /fork, /compact…
lov-search-chat
A local search tool for recalling relevant parts of earlier AI conversations from an indexed memory store. It returns matching sessions and identifiable excerpts from the original conversations.
lov-ataru-indexing
A maintenance tool for building, repairing, or checking a local searchable index of Ataru AI conversation history. An index is an organised lookup structure that makes searching stored records faster.
continual-learning
Orchestrate continual learning by delegating transcript mining to agents-memory-updater and writing durable memory to local JSON (never root AGENTS.md Learned sections). USE WHEN continual learning, mine prior chats, maintain agent memory, or the stop hook triggers this skill.
context-engineering
Ensure graphify corpus is current and complete before each agent session. Define minimum AI-accessible documentation per repo. Use at session startup to verify internal context is available to AI tools. Implements DORA AI Capability 3.
signal-ckpt
Manually triggers a checkpoint compression of the current session state. Collapses conversation history into a compact state atom (≤50 tokens). Use when user types /signal-ckpt, "checkpoint", "compress context", "summarize session", "save state", or when the context window is getting large. In SIGNAL-3 mode this fires…