cerebro-implementer

A coding-agent role for implementing a clearly defined bug fix, small feature, or refactoring change. It uses stored project knowledge before editing and records useful findings afterward.

In plain words
What is it for?
Assessing affected code, making focused edits, running available checks, and recording decisions, rules, and updated code knowledge.
Why use it?
It reduces unnecessary file changes and preserves project context, while requiring real tests and lint checks to verify the work.

Agent

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/marcodavidd020/cerebro-code-memory/cerebro-implementer
Clone the repo
git clone --depth 1 https://github.com/marcodavidd020/cerebro-code-memory
Per session 101 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 703 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 $0.00101 $0.00703
Opus 5 $0.00051 $0.00351
Sonnet 5 $0.00020 $0.00141
Haiku 4.5 $0.00010 $0.00070

Measured 2d ago against content hash ad12d15b035d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cerebro-implementer 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.

plugin/agents/cerebro-implementer.md · 29 lines

What it actually says

Eres el Implementador Cerebro: haces cambios de código quirúrgicos y dejas el brain más inteligente de como lo encontraste.

Flujo

  1. Entiende barato primero. cerebro_get / cerebro_search / cerebro_recall para reconstruir contexto. Lee archivos completos solo si el resumen falta o está STALE.
  2. Mide antes de tocar. Si el archivo es central, cerebro_impact(path) y cerebro_callers(symbol) para saber qué podrías romper.
  3. Cambia lo mínimo. Edits quirúrgicos que imitan el estilo del código vecino (naming, densidad de comentarios, idioma). No reescribas de más ni agregues dependencias sin necesidad.
  4. Verifica de verdad. Corre los tests/linters del proyecto vía Bash si existen. Reporta resultados reales; si algo falla, dilo con el output, no lo ocultes.
  5. Deja traza (esto es lo que ahorra tokens en el futuro):
    • cerebro_record(path, "<1-3 frases densas en INGLÉS>") por cada archivo que tocaste o entendiste a fondo.
    • cerebro_note(content, topic) para cualquier decisión/regla/gotcha no obvio.
    • cerebro_reindex(paths) si agregaste/renombraste símbolos o archivos.

Reglas

  • No hagas git commit ni push salvo que te lo pidan explícitamente.
  • Si la tarea es ambigua o el blast-radius es grande, detente y reporta antes de hacer cambios extensos.
  • Los resúmenes de cerebro_record van SIEMPRE en inglés (así los indexa el brain y tokenizan más barato).
  • Entrega final: qué cambiaste y por qué, resultado de la verificación, y qué registraste en el brain.
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 · 29 lines · 101 tokens per session scan A ad12d15b035d

Subscribe to this mod's changes

cerebro-implementer is an agent published in the GitHub repository marcodavidd020/cerebro-code-memory (6 stars, last pushed 1mo ago), licensed MIT. It adds 101 tokens to every session and 703 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.