ana-learn

A code-quality review agent that checks whether earlier findings still match the current code and records lasting lessons.

In plain words
What is it for?
Use it to recheck reported issues, close findings that are fixed, identify repeated patterns, and turn developer observations into improvements.
Why use it?
It prevents outdated warnings from cluttering the project and helps recurring problems become shared rules. It checks code and history before presenting evidence for the developer's decision.

Agent for Claude Code

Install

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.

agentmods
npx agentmods add agents/anatomia-dev/anatomia/ana-learn
Clone the repo
git clone --depth 1 https://github.com/anatomia-dev/anatomia

Made for: Claude Code.

Per session 20 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 8,259 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00020 $0.08259
Opus 5 $0.00010 $0.04130
Sonnet 5 $0.00004 $0.01652
Haiku 4.5 $0.00002 $0.00826

Measured 2d ago against content hash ee5ccdc02619, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ana-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 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.

.claude/agents/ana-learn.md · 534 lines

How it starts

The opening of the file, as written. The whole thing — 534 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Ana Learn

You are Ana Learn — the quality gardener for this project. You stand between the proof chain and the codebase, asking one question: "is this still true?" Findings make claims about code. Code changes. Claims go stale. Your job is to catch the gap — verify every claim against the current code, close what's resolved, promote recurring patterns into skill rules, and route developer observations into permanent system improvements.

Your disposition is skeptical of the record. The proof chain says a catch block swallows errors silently — does it still? The proof chain says a test uses a weak matcher — was it strengthened? You read the code, check the git history, and answer with evidence. The developer makes the final call. You make the call easy by doing the homework.


Pipeline Position

You are the fifth agent — not in the pipeline, but running alongside it:

Ana → Plan → Build → Verify → proof chain → Learn (you)

The pipeline produces code, tests, and findings. Findings accumulate in the proof chain. You tend the proof chain — closing what's resolved, promoting patterns into skill rules, and helping the developer understand what the findings mean. You run between pipeline sessions, not during them.


On Startup

0. Ground Yourself (MANDATORY — before anything else)

Before responding to the user, before triaging, before doing anything — read these files and run these checks. Every one. No shortcuts.

  1. Branch check. Run git branch --show-current. Read artifactBranch from .ana/ana.json. If you're not on the artifact branch: "You're on {current}. Learn needs {artifactBranch} to execute close and promote commands. Switch now, or triage read-only?" Don't waste a session on triage you can't execute.

  2. Pipeline check. Run ana work status. Note active scopes for context — their findings arrive after verify completes and work complete runs. Triage existing findings normally regardless of pipeline activity. The proof commands handle concurrent access mechanically.

Read the full file on GitHub · 534 lines

Changes

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.

  1. 2d ago First seen · 534 lines · 20 tokens per session scan A ee5ccdc02619

Subscribe to this mod's changes

ana-learn is an agent published in the GitHub repository anatomia-dev/anatomia (32 stars, last pushed 29d ago), licensed MIT. It adds 20 tokens to every session and 8,259 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 2026-08-30.

Related

Other agents, from other repositories

p4s-bugfix-orchestrator

RTL bug fix orchestrator. Manages the full analyze→fix→lint→TB→sim cycle with parallel UNITFIX across modules, Phase 5→4 feedback return, and lesson-learned recording.

babyworm/rtl-agent-team · 52 tokens

lesson-learner

Use proactively when the user asks to encode insights or guidelines into CLAUDE.md. Triggers include phrases like "update your memory", "remember to do this", "learn from this", or "add this to the guidelines". Extracts generalizable principles from context and proposes targeted improvements to project documentation.

CRJFisher/ariadne · 66 tokens

aria

ARIA — AI engineering. New agents, skills, evals, MCP wiring, department growth. Use when designing or changing the crew, when a capability should become a file, or when Josh says teach/grow the department. Use proactively when a one-off prompt is about to become a standing job.

jmenzies722/shua-labs · 62 tokens

mentor

MENTOR — Teaching and growth. Lessons, onboarding, glossary, making the crew easy to learn. Use after a miss, when a new model or person joins, or when Josh says teach the department. Use proactively when a rule lived only in chat.

jmenzies722/shua-labs · 53 tokens

verifier

Verify that work is complete by gathering fresh evidence — run tests, check requirements, confirm claims.

prime-radiant-inc/sprout · 21 tokens

akihirookuda95-openai-agents-sdk-tutorial

Agent "akihirookuda95-openai-agents-sdk-tutorial" from SAIRAMANALADI/vybe-intelligence-vault, covering akihirookuda95/openai-agents-sdk-tutorial, summary, why it matters, repository details and possible use cases.

SAIRAMANALADI/vybe-intelligence-vault · 0 tokens