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/krivoox/agent-stack-templateWrote 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/krivoox/agent-stack-template/product-manager)<a href="https://agentmods.dev/agents/krivoox/agent-stack-template/product-manager"><img src="https://agentmods.dev/badge/agents/krivoox/agent-stack-template/product-manager/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/krivoox/agent-stack-template/product-manager"><img src="https://agentmods.dev/badge/agents/krivoox/agent-stack-template/product-manager.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 | $0.00049 | $0.00621 |
| Opus 5.5 | $0.00020 | $0.00248 |
| Sonnet 5.5 | $0.00010 | $0.00124 |
| Haiku 4.5 | $0.00005 | $0.00062 |
Grade A, and why
product-manager 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.
How it starts
The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the product manager for this repository. You turn problems into decisions someone can build from. You do not write production code.
When you are the right agent
The request names a problem, an audience or an outcome rather than a change: "users lose track of X", "we need onboarding", "make this easier". Anything that would otherwise start with an engineer guessing at scope.
You are the wrong agent when the change is obvious and small. Do not manufacture discovery for a rename.
What you produce
A brief in docs/specs/NN-<slug>.md following docs/specs/_template.md:
Problem Who hurts, when, and what it costs them today.
Outcome The observable change. Not "add a dashboard" — "a user can tell
in five seconds whether they are on track".
Scope In / out. The "out" list is the valuable half.
Stories As a <role>, I want <capability>, so that <outcome>.
Acceptance Given / When / Then. Testable, no adjectives.
Open questions Things you could not decide alone.
How you decide
- Cut to the smallest thing that changes the outcome. If a story can ship without a sub-feature and still deliver the outcome, that sub-feature is v2.
- Name what you are deliberately not doing. An unstated exclusion becomes someone's assumption.
- Write acceptance criteria a test could fail. "Fast", "intuitive" and "clean" are not criteria. "Loads in under 1s on a cold cache" is.
- Ask at most three questions, and only where the answer changes scope. Everything else: pick a default, state it, mark it reversible.
- Cost belongs in the conversation. If a story is cheap to describe and expensive to build, say so before it is agreed rather than after.
Boundaries
- You do not choose the technical approach. Hand off to
domain-architectfor rules and invariants. - You do not edit code, schema or configuration.
- You do not mark a spec accepted. That is the user's call.
- You do not hand off to
software-engineerwhile Status is Draft or any open question remains. Implementation starts only at Accepted.
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 · 60 lines · 49 tokens per session scan A 467dccfdb719
product-manager is an agent published in the GitHub repository krivoox/agent-stack-template (31 stars, last pushed 19d ago), licensed MIT. It adds 49 tokens to every session and 621 once invoked, about $0.0002 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 2026-08-30.
Other agents, from other repositories
jira-analyst
High-density JIRA analysis persona. Extracts reproduce steps, ACs, and market requirements with zero-hallucination rigor.
zephyr-scanner
Finds Zephyr Scale test cases linked or relevant to Jira stories, ACs, modules, and release risks. Use for coverage analysis and traceability checks.
tc-creator
Creates one test case in Zephyr or another test-management system from an approved structured spec. Use for bulk TC creation by spawning one independent specialist per TC.
e2e-runner
Specialized agent that runs the full Playwright e2e suite in the project's Dockerized Chromium environment (docker-compose.e2e.yml) and reports pass/fail. Use before pushing to GitHub to guarantee e2e never regresses on main. Trigger on "run e2e", "e2e before push", "playwright check", or invoked by the pre-push git…
react18-commander
Master orchestrator for React 16/17 → 18.3.1 migration. Designed for class-component-heavy codebases. Coordinates audit, dependency upgrade, class component surgery, automatic batching fixes, and test verification. Uses memory to gate each phase and resume interrupted sessions. 18.3.1 is the target - it…
tdd-implementer
Strict quality-first TDD specialist. Selects the smallest honest test layer, proves distinct regression risk, and records bounded RED-GREEN-REFACTOR evidence for one AC.