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/marcusgoll/spec-flow/routinggit clone --depth 1 https://github.com/marcusgoll/Spec-FlowWhat 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 | $0.00000 | $0.01797 |
| Opus 5 | $0.00000 | $0.00898 |
| Sonnet 5 | $0.00000 | $0.00359 |
| Haiku 4.5 | $0.00000 | $0.00180 |
Grade A, and why
ROUTING 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 yesterday.
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 — 228 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Routing Guide
When to use which agent for implementation work
Quick Decision Tree
Is this a workflow phase (spec, plan, tasks, optimize, ship)?
├── Yes → Use PHASE AGENT (spec-agent, plan-agent, etc.)
└── No → Continue...
Is domain-memory.yaml present?
├── Yes → Use WORKER (domain/worker.md)
│ Worker reads domain memory, picks ONE feature, implements, exits
└── No → Continue...
Is this sprint-level work in an epic?
├── Yes → Use SPECIALIST based on sprint domain:
│ • Backend-only sprint → backend-dev
│ • Frontend-only sprint → frontend-dev
│ • Database sprint → database-architect
│ • Testing sprint → qa-test
│ • Mixed/unclear → general-purpose
└── No → Use WORKER with initialized domain memory
Agent Categories
1. Phase Agents (Workflow Orchestration)
Location: .claude/agents/phase/*-agent.md
When to use: Always spawned by orchestrator commands (/feature, /epic), never directly.
| Agent | Spawned By | Purpose |
|---|---|---|
| spec-agent | /feature, /epic | Generate specification |
| clarify-agent | /clarify | Resolve ambiguities |
| plan-agent | /plan | Create implementation plan |
| tasks-agent | /tasks | Break down into TDD tasks |
| optimize-agent | /optimize | Run quality gates |
| ship-agent | /ship | Handle deployment |
| finalize-agent | /finalize | Archive and document |
Key trait: Return structured results (---COMPLETED---, ---NEEDS_INPUT---, ---FAILED---)
2. Worker Agent (Atomic Implementation)
Location: .claude/agents/domain/worker.md
When to use:
domain-memory.yamlexists in feature/sprint directory- Implementing individual features from a backlog
- Default for
/implementphase
How it works:
- Reads domain-memory.yaml
- Picks ONE failing/untested feature
- Implements with TDD
- Updates domain-memory.yaml
- Commits and EXITs (even if more work remains)
Key trait: Atomic, stateless, disciplined. Never works on more than one feature.
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.
- yesterday First seen · 228 lines · 0 tokens per session scan A 5dfb50344e9c
ROUTING is an agent published in the GitHub repository marcusgoll/Spec-Flow (91 stars, last pushed 4mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,797 tokens. 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
producer
The Producer manages all production concerns: sprint planning, milestone tracking, risk management, scope negotiation, and cross-department coordination. This is the primary coordination agent. Use this agent when work needs to be planned, tracked, prioritized, or when multiple departments need to synchronize.
pm
Use when: checking board status, triaging issues, planning work, deciding what to work on next, organising the backlog. Project Manager — reads the board and helps humans make delivery decisions.
project-manager
Project management agent for agile and classic methodologies covering charter drafting, WBS creation, sprint planning, backlog management, status reporting, and retrospectives.
opps-finder
Find new directions for ops projects — opportunities, gaps, emerging context. Runs /find-opps autonomously, returns backlog items.
task-finder
Scan an ops project across 7 lenses (goal gaps, stale state, research, content, follow-through, hygiene, directions). Updates backlog.
launch-agent
Coordinates the public launch of a vibe-coder SaaS — Product Hunt, Hacker News, social, first-100-users, landing page polish, and post-launch monitoring. Activated when the user says "vamos lançar", "tá pronto pra mostrar", "como divulgo", or when the prototype-lab and sprint-1 are validated.