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 skills/authgear/authgear-server/api-designnpx skills add authgear/authgear-server --skill api-designgit clone --depth 1 https://github.com/authgear/authgear-serverWhat 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 | $0.00040 | $0.02472 |
| Opus 5 | $0.00020 | $0.01236 |
| Sonnet 5 | $0.00008 | $0.00494 |
| Haiku 4.5 | $0.00004 | $0.00247 |
Grade A, and why
api-design 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 2d 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 — 234 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an Authgear API design expert. Based on the user's input, determine the mode:
- Review mode: User provides a specific design draft (e.g., "Review this config struct:", "Here is my proposed GraphQL mutation:"). Evaluate it against the checklist.
- Ideation mode: User describes a feature or idea without a concrete draft (e.g., "Design an API for X", "How should we add Y?"). Develop a design first, then self-review it.
Step 1: Load context (parallel)
Read all of these files in parallel:
docs/specs/glossary.md— canonical termsdocs/specs/convention.md— naming and design conventionsdocs/specs/config.md— config YAML conventionsdocs/specs/api.md— HTTP API conventionsdocs/specs/api-admin.md— Admin GraphQL API conventions
Also read any feature-specific spec that is clearly relevant to the user's input (e.g., if the user mentions "authentication flow", read docs/specs/authentication-flow.md). Read these in the same parallel batch.
Step 2: Load relevant code (parallel)
Based on which API surface is involved, read existing code for pattern comparison. Determine the surface from the user's input, then read the relevant files in parallel:
- Config (
authgear.yaml): Readpkg/lib/config/config.goand one similar feature config file (e.g.,pkg/lib/config/bot_protection.goorpkg/lib/config/fraud_protection.go) - Admin API (GraphQL): Read 2–3 representative files from
pkg/admin/graphql/— pick one node file and one mutation file that are closest to the proposed feature - HTTP API: Read 1–2 handler files from
pkg/auth/handler/api/that are closest to the proposed feature - Authflow API: Read
docs/specs/authentication-flow-api-reference.md - Account Management API: Read
docs/specs/account-management-api.md
If multiple surfaces are involved, load code for all of them.
Step 3: Analyze
Review mode
Evaluate the design against every item in the checklist below. For each item, assign a status:
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.
- 2d ago First seen · 234 lines · 40 tokens per session scan A 58975b084d9d
api-design is a skill published in the GitHub repository authgear/authgear-server (2,017 stars, last pushed today), licensed Apache-2.0. It adds 40 tokens to every session and 2,472 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 skills, from other repositories
auth0-docs
Auth0 — identity platform: auth flows, Universal Login, SSO, identity providers, MFA, RBAC, Actions, tokens.
zeroid
MANUAL TRIGGER ONLY: invoke only when user types /zeroid. Identity infrastructure for AI agents — register identities, issue tokens, delegate to sub-agents, revoke credentials, and manage credential policies via the Zeroid REST API.
zeroid
Identity infrastructure for AI agents — register identities, issue tokens, delegate to sub-agents, revoke credentials, manage policies.
auth-oauth
Implements and reviews OAuth 2.0/2.1 flows following RFC 6749, RFC 9700 (BCP), and PKCE (RFC 7636). Use when implementing authorization code flow, token handling, client registration, or securing OAuth endpoints.
schema
Database schema rules for QAuth. Use when working with the identity model, Drizzle ORM, migrations, repositories, or claim resolution. Reflects the CURRENT shipped schema — the identifier-abstraction model (ADR-002, IMPLEMENTED via Epic.
fastify
Fastify usage for QAuth — plugin authoring (encapsulation, decorators, lifecycle), route schemas with the Zod type provider, production/reverse-proxy setup, and the Fastify version/CVE requirement. Use when adding or modifying @qauth-labs/fastify-plugin- libraries, auth-server routes, or deployment config.