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 skills add mshadmanrahman/pm-pilot --skill orchestratorgit clone --depth 1 https://github.com/mshadmanrahman/pm-pilotWrote 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/skills/mshadmanrahman/pm-pilot/orchestrator)<a href="https://agentmods.dev/skills/mshadmanrahman/pm-pilot/orchestrator"><img src="https://agentmods.dev/badge/skills/mshadmanrahman/pm-pilot/orchestrator.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00013 | $0.00551 |
| Opus 5 | $0.00006 | $0.00275 |
| Sonnet 5 | $0.00003 | $0.00110 |
| Haiku 4.5 | $0.00001 | $0.00055 |
Grade A, and why
orchestrator 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 8d 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orchestrator
Decompose complex work into independent waves of parallel sub-agent tasks.
When to Use
- Tasks with 3+ independent sub-tasks
- Research across multiple sources
- Bulk operations (triage, analysis, review)
- Any work where parallelism saves time
Procedure
-
Decompose Work
- Break task into atomic, independent units
- Group into waves (units within a wave have zero dependencies)
- Order waves by dependency: wave 2 depends on wave 1 results
-
Plan Waves
Wave 1 (parallel): [task-a, task-b, task-c] Wave 2 (parallel, depends on wave 1): [task-d, task-e] Wave 3 (sequential): [merge results] -
Launch Sub-Agents
- One Task agent per wave item
- Each agent gets: clear objective, input data, output format
- Each agent writes results to
.claude/manifests/{task-id}.md
-
Collect Results
- Wait for all agents in a wave to complete
- Read manifests using
manifest-readerskill - Check for failures or partial results
-
Merge and Synthesize
- Combine results from all waves
- Resolve conflicts or inconsistencies
- Produce unified output
Wave Design Rules
- Max 5 agents per wave (resource constraint)
- Each agent should complete in under 2 minutes
- If an agent fails, note failure and continue (don't block the wave)
- Always have a merge step as the final wave
Manifest Format
Each sub-agent writes:
---
task: {task-id}
status: complete | failed | partial
agent: {agent-description}
---
## Findings
{results}
## Errors
{any errors encountered}
Output Format
Orchestration: {task name}
Waves: {count}
Results: {success count}/{total count} tasks completed
Summary: {merged findings}
Failures: {list or "none"}
Rules
- Never launch wave 2 before wave 1 completes
- Always check manifests before proceeding
- Failed agents should not block successful ones
- Keep each agent's scope small and focused
- Clean up manifest files after final merge
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.
- 8d ago First seen · 93 lines · 13 tokens per session scan A 318bc6581e0c
orchestrator is a skill published in the GitHub repository mshadmanrahman/pm-pilot (20 stars, last pushed 2d ago), licensed MIT. It adds 13 tokens to every session and 551 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 skills, from other repositories
build-implementation
Generate context-aware implementation prompts for a selected growth loop. Use when the user says "build", "implement", "generate code", "create prompt", or "how do I build this".
analyze-codebase
Run comprehensive PLG analysis on a codebase to detect tech stack, existing growth features, and revenue opportunities. Use when the user says "analyze", "scan", "audit codebase", or "find growth opportunities".
deploy-telemetry
Set up analytics and tracking infrastructure for growth loops. Use when the user says "deploy telemetry", "set up analytics", "tracking", "events", "push to supabase", or "skene push".
plan-growth-loop
Generate prioritized growth loops with implementation roadmaps based on codebase analysis. Use when the user says "plan", "growth loops", "prioritize", "what should I build", or "roadmap".
validate-loop
Check if growth loop requirements are actually implemented in the codebase. Use when the user says "validate", "check status", "skene status", "is it done", or "verify implementation".
initialize-config
Set up Skene configuration for a project. Use when starting a new PLG analysis, when .skene.config is missing, or when the user says "skene init", "set up skene", or "configure skene".