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/FortiumPartners/ensembleWrote 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/fortiumpartners/ensemble/generate-router-rules)<a href="https://agentmods.dev/commands/fortiumpartners/ensemble/generate-router-rules"><img src="https://agentmods.dev/badge/commands/fortiumpartners/ensemble/generate-router-rules.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 5.1 | $0.00017 | $0.01369 |
| Opus 5 | $0.00009 | $0.00685 |
| Sonnet 5 | $0.00003 | $0.00274 |
| Haiku 4.5 | $0.00002 | $0.00137 |
Grade A, and why
ensemble:generate-router-rules 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 4d 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 — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Generate Router Rules
Introspect available agents and skills in this Claude Code session, then generate
a router-rules.json with categorized keyword mappings for the UserPromptSubmit
routing hook.
Phase 1: Agent Discovery
1.1 Extract Available Agents
Examine your own Task tool documentation to extract all available subagent_type
values. Look in the Task tool description for the list of agent types.
Actions:
- Parse the subagent_type options from the Task tool description
- Extract agent name and description for each
- Note the tools each agent has access to
1.2 Categorize Agents by Domain
Group discovered agents into these functional domains:
| Domain | Description | Example Agents |
|---|---|---|
product_documentation |
PRD, TRD, docs, requirements | product-management-orchestrator, documentation-specialist |
orchestration |
Coordination and planning | ensemble-orchestrator, tech-lead-orchestrator |
development |
Code implementation | frontend-developer, backend-developer, mobile-developer |
quality_testing |
Review, testing, debugging | code-reviewer, test-runner, deep-debugger |
infrastructure_build |
Deploy, build, release | deployment-orchestrator, build-orchestrator |
utility |
Support tasks | general-purpose, file-creator, git-workflow |
database |
Database operations | postgresql-specialist |
Phase 2: Skill Discovery
2.1 Extract Available Skills
Examine the Skill tool documentation's <available_skills> section to extract
all available skills.
Actions:
- Parse skill names from the Skill tool's available_skills list
- Extract description and location for each skill
- Note any argument hints or usage patterns
2.2 Generate Skill Keywords
For each skill, create keyword mappings:
| Keyword Type | Example for vercel skill |
|---|---|
| Primary | vercel, nextjs, next.js |
| Technology | edge function, serverless, preview |
| Action | deploy, preview, domains |
| Diagnostic | vercel issue, vercel problem, not working |
| Pattern | deploy.*react, next.*app, vercel.*issue |
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.
- 4d ago First seen · 175 lines · 17 tokens per session scan A 79a5aa1a94e3
ensemble:generate-router-rules is a command published in the GitHub repository FortiumPartners/ensemble (11 stars, last pushed 1mo ago), licensed MIT. It adds 17 tokens to every session and 1,369 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 2026-09-03.
Other commands, from other repositories
README
Custom commands are reusable prompt templates you invoke with /command-name in Claude Code.
generate-prp
Generate comprehensive PRP (Product Requirements & Plans) with thorough research and validation.
execute-prp
Implement features from PRP specifications with comprehensive validation and testing.
claudetower:status
A status command for ClaudeTower, a command-line tool that adds information to a coding assistant's status line.
setup
Set up codachi as your statusline.
init
Initialize СС Blueprint Toolkit - copy actual documentation templates to your project.