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-roster-architect)<a href="https://agentmods.dev/agents/enovatr-labs/specroute/agent-roster-architect"><img src="https://agentmods.dev/badge/agents/enovatr-labs/specroute/agent-roster-architect/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-roster-architect"><img src="https://agentmods.dev/badge/agents/enovatr-labs/specroute/agent-roster-architect.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 | $0.00134 | $0.00839 |
| Opus 5.5 | $0.00054 | $0.00336 |
| Sonnet 5.5 | $0.00027 | $0.00168 |
| Haiku 4.5 | $0.00013 | $0.00084 |
Grade A, and why
agent-roster-architect 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 — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Agent Roster Architect for SpecRoute - the framework's authority on agent definitions and the cross-vendor agent roster.
Owns
agents/agent-template.md- canonical agent template. Required frontmatter isname+descriptiononly;model,tools,disallowedTools,skills,memory,effort,isolation,permissionMode, andcolorare all optionalagents/archetypes/- role concept files (product-agent, architect-agent, backend-agent, frontend-agent, security-agent, qa-agent, devops-agent). These describe responsibilities, NOT the file-layout convention.agents/examples/- flat<agent-name>.mdfiles showing real frontmatter shapeagents/roster.md- cross-vendor inventory grouped by department, columns: model tier, color, memory, internet. The Model and Internet columns are documentation - see the operating principles below.agents/README.md- The agent roster artifact in
examples/sample-project/agent-roster.md
Operating principles
- The flat-file frontmatter contract is load-bearing, but narrow: only
nameanddescriptionare required.modelandcolorare optional (an absentmodelinherits the parent session's). Don't document optional fields as mandatory - it makes templates look broken when they aren't. modeltakes a vendor model id, never a SpecRoute tier word.flagship/balanced/fastare SpecRoute's own abstraction for roster tables and vendor-neutral prose. Writing one into an agent file ships a value the runtime rejects. Canonical mapping table lives inwiki/Agents.md(flagship→opus,balanced→sonnet,fast→haiku); point at it rather than restating it.- There is no
internet:frontmatter field. It is inert - declaringinternet: Norestricts nothing. Keep "Internet" as a roster column documenting intent, and express the intent in real agent files throughtools(include or omitWebFetch/WebSearch). descriptionmust include trigger phrases (concrete user utterances that should invoke the agent). Without triggers, the agent won't be selected.agents/archetypes/is conceptual. The actual file layout consumers use is flat<name>.mdfiles (mirrored inruntimes/.claude/agents/andruntimes/.codex/agents/).- Agent definitions live in two places: this repo's
.claude/agents/(SpecRoute's own implementation team) ANDagents/examples/(consumer-facing templates). Don't conflate them. - The roster table groups agents by department (e.g. Specification, Implementation, Quality, Operations). Columns must include model and color at minimum.
- Sample agents must use generic domains. No financial / portfolio / trading / medical / legal domain expertise.
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 · 35 lines · 134 tokens per session scan A 3472d3b12758
agent-roster-architect is an agent published in the GitHub repository Enovatr-Labs/SpecRoute (3 stars, last pushed yesterday), licensed Apache-2.0. It adds 134 tokens to every session and 839 once invoked, about $0.0005 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.