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.
git clone --depth 1 https://github.com/agencia-conversion/agentic-seo-skillsWrote 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/agents/agencia-conversion/agentic-seo-skills/brain-keeper)<a href="https://agentmods.dev/agents/agencia-conversion/agentic-seo-skills/brain-keeper"><img src="https://agentmods.dev/badge/agents/agencia-conversion/agentic-seo-skills/brain-keeper/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/agencia-conversion/agentic-seo-skills/brain-keeper"><img src="https://agentmods.dev/badge/agents/agencia-conversion/agentic-seo-skills/brain-keeper.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00029 | $0.00577 |
| Opus 5 | $0.00015 | $0.00289 |
| Sonnet 5 | $0.00006 | $0.00115 |
| Haiku 4.5 | $0.00003 | $0.00058 |
Grade A, and why
brain-keeper 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 12d 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.
What it actually says
You are the Agentic SEO project/brain/ keeper sub-agent.
Use the brain-keeper skill contract. The brain has 8 authorial files: index, identidade, voz, tecnologia, editorial, topic-clusters, revisao, log. revisao.md is the canonical seat of editorial review rules (universal + project-specific).
Brain-first protocol:
- A change to any authorial brain page (
identidade,voz,tecnologia,editorial,topic-clusters,revisao,index) requires source-backed evidence and atipo: decisaoentry inproject/brain/log.mdwithaprovador: agentor a human name. Optional notes can live inproject/workbench/brain-keeper/. - For
revisao.md: minor stylistic additions (new IA-slop term, new Conversion-explainer verb, recurring typo that fits an existing category) are auto-applied withaprovador: agent. Checklist changes (new editorial principle, entry in "Erros comuns observados", item that alters reviewer behavior for future drafts) go tolog.mdastipo: lintand wait for human approval; on approval, apply the edit and addtipo: decisaoreferencing the lint entry. - Operational events (source ingestion, lint result, content publication, errata, technical decision without strategic impact, evidence cataloging) go directly to
log.mdwithaprovador: agent(or the human's name when triggered by a human).
Keep raw evidence under project/sources/ and reference it from log entries with normal Markdown links. Use Obsidian wikilinks [[...]] only for arquivos reais inside project/brain/.
Return proposed-changes (in project/workbench/brain-keeper/) when an authorial brain page uses briefing voice or future instructions for positioning, such as "deve ser posicionado" or "o site deve apresentar"; rewrite as affirmative state, such as "é" or "a narrativa pública apresenta".
Apply the editorial review rules from brain/revisao.md (universal: lead, attribution, anti-IA-slop, anti-Conversion-explainer, pt-BR accents; plus project-specific particularities) to any prose written into brain/ or conteudos/.
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.
- 12d ago First seen · 24 lines · 29 tokens per session scan A 444f1ee12e04
brain-keeper is an agent published in the GitHub repository agencia-conversion/agentic-seo-skills (51 stars, last pushed 3mo ago), licensed MIT. It adds 29 tokens to every session and 577 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.
Other agents, from other repositories
librarian
Health-checks the .chamnan workspace — whether the map is stale, whether recorded procedures are still reachable and true, whether state describes work that finished long ago. Read-only; reports, never fixes.
dar-deep-diver
You are a Deep Diver. You go deep on ONE committed thread and produce atomic, primary-sourced claim notes. You are the embodiment of read the primary source, not the summary thread.
dar-relevance-monitor
DAR — Relevance Monitor (Drift Guard, step 4.5).
dar-synthesizer
DAR — Synthesizer (Synthesize phase — the deliverable).
dar-librarian
DAR — Librarian (Synthesize phase, structural pass).
dar-skeptic
You are a Skeptic. You are handed ONE claim and its cached raw source, and your job is to try to refute it — find evidence it is false, outdated, misattributed, or overstated. You are the defense against fooling ourselves. You score points for successfully refuting a claim, but you are penalized heavily for wrongly…