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/Enovatr-Labs/SpecRouteWrote 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/enovatr-labs/specroute/agent-template)<a href="https://agentmods.dev/agents/enovatr-labs/specroute/agent-template"><img src="https://agentmods.dev/badge/agents/enovatr-labs/specroute/agent-template/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/agents/enovatr-labs/specroute/agent-template"><img src="https://agentmods.dev/badge/agents/enovatr-labs/specroute/agent-template.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 | $0.00080 | $0.01147 |
| Opus 5.5 | $0.00032 | $0.00459 |
| Sonnet 5.5 | $0.00016 | $0.00229 |
| Haiku 4.5 | $0.00008 | $0.00115 |
Grade A, and why
<slug-lowercase-hyphenated> 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 2mo 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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the for - <one-sentence statement of the agent's authority and scope>.
Owns
<path/to/files>- <what's there and why this agent owns it><path/to/files>- <…><path/to/files>- <…>
List concrete file and directory ownership. Ownership is exclusive: two agents owning the same files is a coordination tax. Use the Don't use for section to hand off neighbors.
Operating principles
- <Principle 1: a non-obvious rule the agent should follow. Be specific.>
- <Principle 2.>
- <Principle 3.>
- <Principle 4: usually a "shape" rule, e.g. "every requirement has a stable ID" or "tasks must back-reference requirements".>
- <Principle 5: usually a "scope" rule, e.g. "sample content uses generic domains only".>
These principles are the agent's brief. They override generic best-practice when in conflict.
Don't use for
- <Adjacent concern 1> - .
- <Adjacent concern 2> - .
- <Adjacent concern 3> - .
Crisp boundaries here are how rosters stay healthy. If you find yourself violating "Don't use for," stop and hand off.
How to fill this template
- Pick a slug. Lowercase, hyphen-separated. The filename matches the slug (e.g.
prd-author.md). - Write the description with triggers. This is the most important field. Include 3–5 concrete trigger utterances in quotes - these drive automatic agent selection. Without triggers, the agent is invisible.
- Pick a model - or don't.
nameanddescriptionare the only required fields; every other field below is optional. Whenmodelis absent, Claude Code subagents inherit the parent session's model, which is often the right answer. If you do set it, write a value your runtime accepts (Claude:opus/sonnet/haiku/fable/ a full model id /inherit). SpecRoute'sflagship/balanced/fasttiers are a documentation abstraction for rosters only - never write them into an agent file. Mapping table:wiki/Agents.md. - Decide on web access via
tools. There is nointernet:field - it is not read by any runtime. To let the agent research, includeWebFetch/WebSearchintools; to deny it, omit them from the allowlist or name them indisallowedTools. - Pick a color, optionally. Used by the runtime UI only. Nothing breaks if you omit it.
- List
Owns. Be specific about file paths and directories. The agent's claim to those files is exclusive. - Write 4–6 operating principles. Specific. Non-obvious. The kind of guidance you'd give a senior contributor on day one.
- Write 2–4 "Don't use for" lines. These hand off the adjacent concerns to the right neighbor.
- Body length. ~30–50 lines is the sweet spot. Longer agents tend to drift; shorter agents under-specify.
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.
- 2mo ago First seen · 60 lines · 80 tokens per session scan A 7ebee34ad0e1
<slug-lowercase-hyphenated> is an agent published in the GitHub repository Enovatr-Labs/SpecRoute (3 stars, last pushed yesterday), licensed Apache-2.0. It adds 80 tokens to every session and 1,147 once invoked, about $0.0003 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 2026-08-31.
Other agents, from other repositories
bootstrap
Use once per invocation, at the start of a new Hedgehog project, to land the workspace for whichever core planner selected at Phase 0 — the first real workspace this project gets, since init with no explicit core flag lands the shared agents/skills/build-graph payload and leaves the workspace to bootstrap. Reads the…
doc-onboard
Discovery and understanding of a codebase. Use for a new developer joining the project, to document the architecture, or to understand an open source project.
Software Engineer
Implements features, fixes bugs, and writes production-quality code.
ops-monitoring
Application instrumentation and monitoring, and deploying the observability stack. Use to add structured logs, Prometheus metrics, OpenTelemetry traces, or to stand up Prometheus/Grafana/Loki/Alertmanager.
ops-deps
Audit and analysis of dependencies. Use to check vulnerabilities, identify outdated packages, or plan updates.
refuter-reproduction
Reproduction-first lens for bug-fix DoDs only — the original bug must be reproduced against pre-fix code before any fix claim is trusted. Read-only. Invoked only by mnm-verify, only when the DoD's Goal is a bug fix.