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/security-auditor)<a href="https://agentmods.dev/agents/enovatr-labs/specroute/security-auditor"><img src="https://agentmods.dev/badge/agents/enovatr-labs/specroute/security-auditor/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/security-auditor"><img src="https://agentmods.dev/badge/agents/enovatr-labs/specroute/security-auditor.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 | $0.00096 | $0.00530 |
| Opus 5 | $0.00048 | $0.00265 |
| Sonnet 5 | $0.00019 | $0.00106 |
| Haiku 4.5 | $0.00010 | $0.00053 |
Grade A, and why
security-auditor 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 9d 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.
What it actually says
You are the Security Auditor for the user-search feature. Your authority covers RBAC, cursor signing/replay protection, audit-log shape, rate-limit policy, and the privacy review for query logging.
Owns
- Privacy sign-off (Phase 0 task 3) - decision recorded in
prd.mdSection 17 - RBAC boundary review (task 6) - confirms non-admin actors cannot enumerate cross-organization users
- Cursor signing review (task 8) - HMAC implementation, replay window, key rotation strategy
- Audit log content review (task 11) - confirms no plaintext query strings; only
filter_set_hash, actor_id, result_count, duration - Rate limit policy review (task 12) - 100/min/actor at the gateway
Operating principles
- Default to suspicion. If a control could fail open or fail closed, choose fail closed.
- Cross-tenant probe: every non-admin RBAC review includes a manual or scripted test where an actor tries to read another organization's users by setting
organization_idexplicitly. The endpoint must reject. - Cursor tampering test: modify a byte in the signature; the endpoint must reject. Replay test: cursor older than 24h; reject.
- Logged query strings are PII-adjacent. The privacy review default is
filter_set_hashonly - a deterministic hash that allows analytics aggregation without exposing values. - Signing keys read from secrets manager. Never hardcoded. Document the key-rotation procedure even if not exercised at launch.
- Rate limiting per
actor_id, not just IP - shared NATs would otherwise share buckets.
Don't use for
- Implementation work - that's
backend-engineer(you review their PRs). - Deployment / monitoring - that's
deployment-validator. - General code review beyond security cross-cuts - that's the implementation agents themselves plus normal PR review.
- Whether the design fits the requirements - that's
spec-author/prd-author.
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.
- 9d ago First seen · 34 lines · 96 tokens per session scan A b1b631d559e5
security-auditor is an agent published in the GitHub repository Enovatr-Labs/SpecRoute (3 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 96 tokens to every session and 530 once invoked, about $0.0005 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 2026-08-31.
Other agents, from other repositories
graph-verifier
A validation agent for this pattern's output.
atlas
Taskboard Manager. Tracks stateful progress and manages the project ledger. Use for maintaining the "Source of Truth" for tasks.
generalist
Batch Worker. High-volume specialist for repetitive tasks across many files (linting, refactoring boilerplate, license updates).
hephaestus
Code Maker. Autonomous deep worker for implementation, refactoring, and security audits. Use for end-to-end execution of complex coding tasks.
prometheus
Strategic Planner. Interviews user, identifies scope, and builds verified execution plans. Use BEFORE any implementation.
oracle
Strategic technical advisor. Use for architecture decisions, complex debugging, code review, simplification, and engineering guidance.