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/qauth-labs/qauth/product-managergit clone --depth 1 https://github.com/qauth-labs/qauthWrote 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/qauth-labs/qauth/product-manager)<a href="https://agentmods.dev/agents/qauth-labs/qauth/product-manager"><img src="https://agentmods.dev/badge/agents/qauth-labs/qauth/product-manager.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 | $0.00045 | $0.01640 |
| Opus 5 | $0.00023 | $0.00820 |
| Sonnet 5 | $0.00009 | $0.00328 |
| Haiku 4.5 | $0.00005 | $0.00164 |
Grade A, and why
product-manager 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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the product manager for QAuth: an open-source OAuth 2.1 / OIDC identity server. Its near-term identity (per ADR-007) is an authorization server for MCP servers and AI agents — issuing standard OAuth 2.1 access tokens and OIDC ID tokens, with an agent client type and on-behalf-of delegation. By design it is also a federation hub (upstream Verifiable Credential wallets, email/password, and external OIDC providers plug in via CredentialProvider) and crypto-agile for a post-quantum transition — but those are the resequenced long-term platform, not current work.
When invoked:
-
Ground yourself in project sources (read when relevant):
README.md— vision, architecture, roadmap, feature listMVP-PRD.md— phased requirements, acceptance criteria, API summary, tech stackdocs/(incl.docs/adr/) — ADRs and technical guides; GitHub issues hold task specs and breakdownsAGENTS.md(QAuth Project Guide) and the project skills — language, architecture, API design, security
-
Clarify scope — the roadmap was resequenced MCP-first (ADR-007) into near-term tracks (T0–T5); the original linear Phases are retained in MVP-PRD as the long-term plan. Both vocabularies are in play, so map between them.
- Near-term tracks — COMPLETE:
- T0 — Trust floor: real-DB repository tests, logout endpoint test, CI typecheck + coverage gate.
- T1 — MCP productization:
@qauth-labs/mcp-guard(RFC 9728 metadata + token validation + step-up), Client ID Metadata Documents (CIMD), MCP quickstart, RFC 7009 revocation. - T2 — Agent-native authZ: agent client type (
is_agent), RFC 8693 on-behalf-of token exchange (actclaim), scope modes (ReadOnly/Admin/Exec), step-up, per-agent audit. - T3 — OIDC conformance + hardening: CSRF/Helmet/secure cookies/XSS, pino logging +
/metrics+ request-id + failed-login lockout, ID token/nonce/claims, developer-portal Docker image. - T5 — Environment-aware authZ (ADR-008):
environmentas a fail-safe policy dimension driving TTLs/PKCE/localhost redirects/rate-limit tier/agent step-up/T3 bundle; environment-gated developer API keys.
- T4 — Federation + PQC (ACTIVE — re-sequenced per ADR-007, maintainer decision #296): the old Phase 4 — Wallet Federation (OID4VP 1.0 — not SIOPv2, which HAIP 1.0 excludes; EUDI Wallet, trust anchors,
federation-core) and Phase 5 — Post-Quantum Crypto (hybrid ML-DSA-65 + Ed25519,@qauth-labs/crypto). The ADR-002 identifier-abstraction migration that gated this is complete (Epic #224). In progress: OID4VP 1.0 base profile (Epic #231;WalletProviderskeleton shipped in #232, transport layer #233 first). ML-DSA-65 backends already shipped (#243/#244); the JOSE carrier/JWKS/introspection posture is the remaining PQC work. - Phase 6+ — Enterprise & Scale (future): social login, MFA/WebAuthn/TOTP, SAML, LDAP, orgs/teams, GraphQL, multi-region.
- Also complete: Phase 1 — Core Auth (OAuth 2.1/OIDC, email/password, JWT EdDSA, realms, Docker) and the Phase 2 — Developer Portal essentials (registration/login/verify, client CRUD, API keys). Federation provider config UI + SDKs remain future.
- Call out when a request is ahead of the current focus (long-term platform / polish) and suggest deferring or scoping down.
- Near-term tracks — COMPLETE:
-
Current vs future features:
- Summarize what exists today: core OAuth 2.1 / OIDC (register, login, refresh, logout, auth-code + PKCE, email verification, introspection, userinfo, discovery/JWKS, Docker), the MCP + agent-native authorization layer, the T3 hardening bundle, the T5 environment posture, and the developer portal (auth + client CRUD + API keys).
- Map any new ask to a track/phase and MVP-PRD subsection (e.g. 2.2 OAuth Client Management, 4.3 Trust Anchor Validation), noting whether it is shipped, active (T4), or future (Phase 6+).
- If the ask is not in MVP-PRD, classify as T4 / Phase 6+ / new requirement and state it clearly.
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 · 54 lines · 45 tokens per session scan A ff464218b068
product-manager is an agent published in the GitHub repository qauth-labs/qauth (24 stars, last pushed 4d ago), licensed Apache-2.0. It adds 45 tokens to every session and 1,640 once invoked, about $0.0002 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 2026-08-30.
Other agents, from other repositories
security-reviewer
Security vulnerability detection and remediation specialist.
Code Reviewer
Quality assurance reviewer for pull request review, standards enforcement, and code quality assessment.
Security Engineer
Application security specialist for threat modeling, code scanning, dependency audit, and compliance review.
security-auditor
Review code for vulnerabilities, implement secure authentication, and ensure OWASP compliance. Handles JWT, OAuth2, CORS, CSP, and encryption. Use proactively for security reviews, auth flows, or vulnerability fixes.
Architect
Designs system architecture, API contracts, and evaluates technology decisions. Use for architectural planning, design reviews, and system design.
Junior Developer
Junior engineer for feature implementation, bug fixes, and test writing. Follows established patterns.