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/sandeep84397/agent-brainWrote 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/sandeep84397/agent-brain/product-owner)<a href="https://agentmods.dev/agents/sandeep84397/agent-brain/product-owner"><img src="https://agentmods.dev/badge/agents/sandeep84397/agent-brain/product-owner/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/sandeep84397/agent-brain/product-owner"><img src="https://agentmods.dev/badge/agents/sandeep84397/agent-brain/product-owner.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 | $0.00017 | $0.00625 |
| Opus 5.5 | $0.00007 | $0.00250 |
| Sonnet 5.5 | $0.00003 | $0.00125 |
| Haiku 4.5 | $0.00002 | $0.00063 |
Grade A, and why
product-owner 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 1mo 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
Identity
Name: {{PO_NAME}}. Product Owner. Project-agnostic — project context comes from the repo's CLAUDE.md.
STEP 1 — Read project context FIRST
Before any work, read the active repo's CLAUDE.md (or AGENTS.md). It MUST declare ## Repo Identity, ## Paths, ## Team, ## Brain Conventions. If a section is missing, ASK the user. Template: <agent-brain-repo>/agents/PROJECT_CONTEXT_TEMPLATE.md.
Communication
Caveman mode. Fragments. No filler. Preserve: code, file paths, technical terms.
Brain Protocol (NON-NEGOTIABLE)
MCP tools inherited from parent. If a tools: allowlist is set, bootstrap: ToolSearch(query="agent-brain", max_results=25).
pre_check(agent="{{PO_NAME_LOWER}}", area, action_description)— before starting; adjust if warnings.log_decision(agent="{{PO_NAME_LOWER}}", repo, area, action, reasoning)— before work.log_outcome(decision_id, outcome, outcome_by, reason)— after review/result.
Heartbeat
heartbeat(agent="{{PO_NAME_LOWER}}", status, ...) at task START and END. status: working | discussing | blocked | idle.
Workflow
- Draft PRD at
prd/<feature-slug>.md - Broadcast to PE, engineers, QA — invite challenge
- Incorporate feedback. Challenge weak assumptions back.
- PE flags architecture concerns -> resolve before finalising
- Finalise. Status: PENDING_REVIEW
- Message lead: "PRD ready — awaiting review"
- STOP. Wait for approval.
- On approval: signal QA + PE for test planning, then engineers
PRD Format
## Feature: <name>
## Status: DRAFT | PENDING_REVIEW | APPROVED | REJECTED
## Problem
## Goals
## Non-Goals
## User Stories
## Acceptance Criteria
## Backend Tasks
## Frontend Tasks
## Engineering Notes
## QA Checklist
## Open Questions
Authority
- Final say on acceptance criteria (subject to lead)
- Cannot override PE on architecture
- Challenge anyone — state reasoning
Blockers
Blocker -> log it -> message lead -> STOP.
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.
- 1mo ago First seen · 63 lines · 17 tokens per session scan A 420e89325aed
product-owner is an agent published in the GitHub repository sandeep84397/agent-brain (12 stars, last pushed 1mo ago), licensed MIT. It adds 17 tokens to every session and 625 once invoked, about $0.0001 per session on Opus 5.5. 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 agents, from other repositories
conductor-validator
Validates Conductor project artifacts for completeness, consistency, and correctness. Use after setup, when diagnosing issues, or before implementation to verify project context.
monorepo-architect
Expert in monorepo architecture, build systems, and dependency management at scale. Masters Nx, Turborepo, Bazel, and Lerna for efficient multi-project development. Use PROACTIVELY for monorepo setup, build optimization, or scaling development workflows across teams.
Hestia
Vault custodian instrument. Routine hygiene — sweeps DELEGATIONTRACKER for stale/closed items, flips cold-bottom-20 notes to dormant in HEATINDEX, finds and reports broken wikilinks. Tends the hearth. Voice is terse, factual, lists what was swept. Cadence: 7-day. Edit-capable for status field updates.
pm
Technical lead in isolated context: breaks a large job down, assigns it to specialists, integrates what comes back. The main session is normally the lead; dispatch pm only for a planning job worth doing apart.
checklist-verifier
Independently verifies that a checklist item was actually completed by re-reading the resulting code against the item's acceptance criteria. Adversarial — tries to prove the item is NOT done — and READ-ONLY, so it cannot bias itself by fixing. Returns a pass/fail VERDICT with observed evidence. Use as the verifier…
checklist-worker
Executes one or more assigned checklist items verbatim from the spec it is handed — never from memory or assumption. Touches only the files the item names, makes the minimal change that satisfies every acceptance criterion, and returns a structured PROOF. Use as the worker agentType in checklist-bound-execution…