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/VandanaAjayDubey111/great-pmWrote 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/commands/vandanaajaydubey111/great-pm/pm-help)<a href="https://agentmods.dev/commands/vandanaajaydubey111/great-pm/pm-help"><img src="https://agentmods.dev/badge/commands/vandanaajaydubey111/great-pm/pm-help/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/commands/vandanaajaydubey111/great-pm/pm-help"><img src="https://agentmods.dev/badge/commands/vandanaajaydubey111/great-pm/pm-help.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 | $0.01304 |
| Opus 5.5 | $0.00007 | $0.00522 |
| Sonnet 5.5 | $0.00003 | $0.00261 |
| Haiku 4.5 | $0.00002 | $0.00130 |
Grade A, and why
pm-help 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 26d 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 — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the great-pm /pm-help command. Print every available command,
grouped sensibly. If --brief, print just the command names without
descriptions.
Output
great-pm — your product team in software
THE 6-STAGE LOOP
/grill-me "<idea>" grill-me interrogates YOU — short adaptive rounds
that widen a fuzzy idea and surface unknowns before
you build on it. Works at ANY stage (new initiative,
mid-build feature, pivot, post-launch). Ends in a
situation brief + offer to /pm-start from it.
/pm-start "<problem>" Bootstrap a new initiative. Always offers a grill
first, then auto-runs the Discover
stage and proceeds through to gate:strategy.
/pm-discover <slug> Stage 1: user-researcher + feedback-synthesizer +
market-analyst (parallel). Outputs discovery brief +
competitive brief.
/pm-strategize <slug> Stage 2: product-strategist (+ pricing-strategist with
--with-pricing). Outputs strategy doc with falsifiable
bet + kill criterion.
/pm-prioritize <slug> Stage 3: prioritization-analyst (RICE/WSJF/Kano/MoSCoW)
+ tradeoff-arbiter. Outputs ranked backlog.
/pm-spec <slug> Stage 4: spec-writer + spec-reviewer (mandatory clarity
check). Outputs PRD ready for gate:spec.
/pm-launch <slug> Stage 5: launch-manager + gtm-strategist (parallel).
Outputs launch plan + GTM plan for gate:launch.
/pm-measure <slug> Stage 6: analytics-analyst + experiment-designer.
Outputs performance read-out + next-cycle questions.
The loop closes here.
DECISIONS & GATES
/pm-inbox Show pending gates and decisions waiting on you.
/pm-board Open the project board.
/pm-promote <draft> Promote a draft to a gate-ready Beads issue. Runs
pm-reviewer first; PASS is required.
/pm-review <artefact> Trigger pm-reviewer on any artefact. Returns verdict
unedited. Non-promoting.
/pm-gate approve <id> Explicitly approve a filed gate (great-pm default).
/pm-gate reject <id> <reason> Reject a gate with reason; halts pipeline.
/pm-gate list Show open gates (minimal view; /pm-inbox is richer).
/pm-gate show <id> Full gate state + linked draft + REVIEW doc.
/pm-ownership Show who owns what across initiatives, drafts, gates.
SPECIALIST WORK
/pm-experiment <hypothesis> Author a pre-registered A/B / multivariate / holdout /
switchback experiment.
/pm-metrics <slug> Author the metrics plan BEFORE the build. North Star +
KPIs + instrumentation events.
/pm-roadmap Generate or refresh the roadmap (Now / Next / Later).
/pm-feedback <source> Synthesize a feedback batch into JTBD-clustered themes.
/pm-competitive <target> Competitive teardown — single competitor or whole
landscape around a JTBD.
/pm-pricing <scope> Author or refresh a pricing plan with willingness-to-
pay rationale per audience.
ANALYSIS & AUDIT
/pm-audit [scope] 15-dimension PM-health audit. Severity-rated findings.
/pm-cost Show great-pm token / agent-invocation cost over time.
/pm-burn Show recent agent burn — who ran, how long, what
verdict.
/pm-digest One-screen digest of the last N sessions.
/pm-crystallize <topic> Crystallize a recurring discussion into a permanent
decision doc.
/pm-rfc <topic> Draft an RFC for a non-trivial decision (architecture,
governance, scope change).
/pm-poc <hypothesis> Spec a minimal proof-of-concept that would falsify
the hypothesis cheaply.
SESSION
/pm-save Full end-of-session hand-off (logs + status + lessons).
/pm-learn Lightweight: extract lessons from this session only.
/pm-resume Resume work from the most recent .great-pm/HANDOFF.md.
/pm-doctor Health check on .great-pm/ state — surface drift,
missing files, orphan drafts.
META
/pm-agent-review <agent> Review a great-pm agent's recent verdict distribution
and drift signals.
/pm-agent-retire <agent> Deprecate an agent gracefully (governance-gated, never
deletes).
/pm-help This help.
CONCEPTS
The 6-stage loop Discover → Strategize → Prioritize → Define →
(handoff to engineering / engineering) → Launch → Measure.
The 3 gates gate:strategy, gate:spec, gate:launch — every gate package
passes through pm-reviewer first; the verdict travels to
you unedited.
Parallel-by-dependency Multiple initiatives run in parallel; unrelated work never
blocks unrelated work.
Governance Agents DRAFT and PROPOSE. You DECIDE. One narrow exception:
skill-scout may swap a skill autonomously when isolated and
non-critical (logged to .great-pm/skill-swaps.log).
For more, see ~/great-pm/README.md.
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.
- 26d ago First seen · 117 lines · 17 tokens per session scan A df9e8f380c6e
pm-help is a command published in the GitHub repository VandanaAjayDubey111/great-pm (3 stars, last pushed 26d ago), licensed MIT. It adds 17 tokens to every session and 1,304 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 2026-09-21.
Other commands, from other repositories
rfc
RFC process for cross-team decisions. Create, track, and close RFCs. Accepted RFCs auto-create ADRs.
exception
Signed gate-exception registry — replace ad-hoc --admin / --no-verify bypasses with an auditable, signed, expiring record (who · why · gate · scope · expiry). Create / list / check exceptions.
hatch3r-onboard
Generate a comprehensive onboarding guide for a new developer joining the project -- spawn parallel researchers to analyze codebase structure, architecture, and conventions, then produce a tailored onboarding document with setup instructions, architecture walkthrough, coding conventions, key workflows, tribal…
new-track
Create a new development track with spec, plan, and metadata.
dashboard
Starts the live web dashboard for the current fellowship — quest/scout progress, gate approvals, and event history — in the background, and prints the URL.
explainme
Explain a codebase, subsystem, or PR to a human — visually and verifiably. Grounds a diagram in codegraph's real structure and delivers a self-contained interactive HTML map with guided walkthrough chapters.