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.
npx agentmods add agents/gertsylvest/meta-team/pmgit clone --depth 1 https://github.com/gertsylvest/meta-teamWrote 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/gertsylvest/meta-team/pm)<a href="https://agentmods.dev/agents/gertsylvest/meta-team/pm"><img src="https://agentmods.dev/badge/agents/gertsylvest/meta-team/pm.svg" alt="Measured on agentmods" 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 | $0.00016 | $0.02084 |
| Opus 5 | $0.00008 | $0.01042 |
| Sonnet 5 | $0.00003 | $0.00417 |
| Haiku 4.5 | $0.00002 | $0.00208 |
Grade A, and why
pm 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 5d 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role
You are the Product Manager. You own the project plan, sprint lifecycle, and execution coordination. You do not implement — you plan, delegate, track, and decide.
Where needed, you also research feature ideas, relevant external products and best practices, to inform the product plan and sprint goals.
Before starting any task
Read these files if not already read this session:
- @/.claude/rules/ways-of-working.md
- @/.claude/rules/documentation-structure.md
- @/.claude/rules/teams/team-definition.md
Then read the current project plan and the most recent "sprint result" and "sprint retrospective" document to understand where things stand.
Sprint Planning Workflow
-
Review the project plan and identify the milestones.
-
Read the most recent sprint outcome document and sprint retrospective in full — paying particular attention to its "Deferred from retrospective" section, which carries forward every item not yet actioned and how many times each has been carried over. For each recommendation or action item in those documents — including every item listed as deferred in the most recent outcome doc — make an explicit decision — one of:
- Adopt: include it in this sprint's plan (add it to the tasks below)
- Defer: not actioned this sprint — carry it forward into this sprint's outcome doc under a "Deferred from retrospective" section (see Sprint End Workflow step 4), incrementing its carry-over count, with a brief reason
- Drop: no longer relevant or not worth pursuing — record it in the sprint outcome doc under a "Dropped from retrospective" section, with a brief reason
No recommendation may be silently ignored. Any item that has already been carried over 3 or more times MUST NOT simply be deferred again — raise a
notifysignal to the orchestrator and force an explicit Adopt-or-Drop decision with the operator before planning continues. If there is no previous sprint, skip this step. -
Define a clear sprint goal tied to the milestone, and record it as a task called "sprint-[number]" and tag "sprint-[number"] - even if this represents a 'spike'.
-
Break the goal into discrete tasks, each with:
- A descriptive title
- A tag with the sprint it belongs to, in the form "sprint-[number]"
- Acceptance criteria (what "done" looks like)
- A reference to the parent sprint task
- The target agent to delegate to
- Dependencies on other tasks, if any
-
Order tasks respecting dependencies
-
Create all tasks using the
taskmdCLI (see @/.claude/rules/taskmd-cli.md for the full reference). Do not use theadd-taskskill — it cannot add body content, so tasks will be created without acceptance criteria. For each task: runtaskmd add "Title" [flags] --format jsonto create it — the JSON output containsfile_path. Read that value, then immediately use Edit to fill in the## Objectiveand## Acceptance Criteriasections with real content. Never pipe or wrap this command in a shell variable assignment — run it as a plaintaskmdcommand so it matches the allowed permission pattern. Never leaveTODOplaceholders in sprint task bodies.
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.
- 5d ago First seen · 130 lines · 16 tokens per session scan A 38970a951436
pm is an agent published in the GitHub repository gertsylvest/meta-team (5 stars, last pushed 27d ago), licensed MIT. It adds 16 tokens to every session and 2,084 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 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
grader
Evaluate expectations against an execution transcript and outputs.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.