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/caipe-io/ai-platform-engineeringWrote 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/rules/caipe-io/ai-platform-engineering/specify-rules)<a href="https://agentmods.dev/rules/caipe-io/ai-platform-engineering/specify-rules"><img src="https://agentmods.dev/badge/rules/caipe-io/ai-platform-engineering/specify-rules/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/rules/caipe-io/ai-platform-engineering/specify-rules"><img src="https://agentmods.dev/badge/rules/caipe-io/ai-platform-engineering/specify-rules.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 | $0.01906 | $0.01906 |
| Opus 5.5 | $0.00762 | $0.00762 |
| Sonnet 5.5 | $0.00381 | $0.00381 |
| Haiku 4.5 | $0.00191 | $0.00191 |
Grade A, and why
specify-rules 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 1mo 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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ai-platform-engineering-fix-thinking-collapse Development Guidelines
Auto-generated from all feature plans. Last updated: 2026-05-20
Active Technologies
- N/A (UI state only) (095-fix-thinking-panel-expand)
- Python 3.11+ (supervisor,
skills_middleware), TypeScript / Next.js 16 + React 19 (UI) + LangGraph, LangChain,deepagents(≥0.3.8,SkillsMiddleware), FastAPI (skills_middleware.router), A2A binding (AIPlatformEngineerA2ABinding), httpx, PyJWT/jose for JWKS (097-skills-middleware-integration) - MongoDB (
agent_skills, optionalskills,skill_hubs,catalog_api_keys,skill_scan_findings); filesystemSKILLS_DIRfor packaged defaults; in-process catalog cache with explicit generation counters (097-skills-middleware-integration) - Python 3.11+ (supervisor,
skills_middleware); TypeScript / Node 20+ (Next.js UI) + LangGraph, LangChain,deepagents(≥0.3.8,SkillsMiddleware), FastAPI, A2A protocol; Next.js 16, React 19, Tailwind; optionalcisco-ai-skill-scannerCLI/package for hub/CI scans per skill-scanner (097-skills-middleware-integration) - Python 3.11+ + LangGraph, deepagents, LangChain, FastAPI (098-unify-single-distributed-binding)
- MongoDB/Redis (checkpointer/store — unchanged by this feature) (098-unify-single-distributed-binding)
- Python 3.11+ + LangGraph, LangChain, FastAPI, A2A Protocol, deepagents (≥0.3.8) (098-unify-single-distributed-binding)
- MongoDB (checkpoints, store), Redis (alternative checkpointer), InMemorySaver (fallback) (098-unify-single-distributed-binding)
- Python 3.11+ (supervisor, agents, slack-bot); TypeScript / Next.js (CAIPE UI, BFF API routes) + NextAuth.js (OIDC), MongoDB (users, policies, app metadata), RAG server (existing group-based roles), Slack Bolt (or equivalent) for Slack surfaces (098-enterprise-rbac-slack-ui)
- Hybrid — Keycloak for authz policies (resources, scopes, permissions), realm roles, user attributes (slack_user_id); MongoDB for team/KB assignments, app metadata, operational state (FR-023) (098-enterprise-rbac-slack-ui)
- TypeScript (Next.js 16, React 19) + NextAuth.js, Keycloak Admin REST API, MongoDB (teams), Tailwind CSS, shadcn/ui components (098-enterprise-rbac-slack-ui)
- Keycloak (users, roles, user attributes, sessions) + MongoDB (team memberships, slack_user_metrics) (098-enterprise-rbac-slack-ui)
- TypeScript (Next.js 16, React 19) + Python 3.11+ (Slack bot) + NextAuth.js, Keycloak Admin REST API, MongoDB, Slack Web API (
@slack/web-api), Tailwind CSS, shadcn/ui (098-enterprise-rbac-slack-ui) - Keycloak (users, roles, attributes, sessions) + MongoDB (teams, slack_link_nonces, slack_user_metrics) (098-enterprise-rbac-slack-ui)
- Python 3.11+ (supervisor, agents, Slack bot, RAG server), TypeScript / Next.js 16 + React 19 (UI) + LangGraph, LangChain, FastAPI, NextAuth.js, Keycloak (OIDC + AuthZ Services), Agent Gateway, MongoDB, cel-python (Python CEL), cel-js (TypeScript CEL), Slack Web API (
@slack/web-api), Tailwind CSS, shadcn/ui (098-enterprise-rbac-slack-ui) - MongoDB (teams, KB ownership, app metadata, nonces, metrics), Keycloak (policies, resources, roles, user attributes) (098-enterprise-rbac-slack-ui)
- TypeScript 5.x, Node 20+, Next.js App Router (16), React 19 + Next.js route handlers, Motor-backed patterns via existing Mongo helpers in UI (
getCollection), JWT/session auth via existing API middleware (fix/skills-ai-generate-use-dynamic-agents) - MongoDB collection
agent_skillsonly (no new collection); chart-mountedSKILL.md+metadata.jsonfor packaged templates (fix/skills-ai-generate-use-dynamic-agents) - TypeScript / Next.js 16 + React 19 for CAIPE Admin UI and BFF APIs; Python 3.11+ for Slack bot and AgentGateway OpenFGA bridge. + NextAuth.js, MongoDB driver, Keycloak Admin REST, OpenFGA HTTP API, AgentGateway ext_authz, Slack Web API/Bolt, React Flow for graph visualization. (2026-05-11-identity-group-rebac)
- MongoDB for sync rules, sync runs, external group links, membership sources, team/channel/resource intent, audit metadata, and policy ownership metadata; OpenFGA for authorization tuples; Keycloak for identity, token issuance, limited realm roles, users, and upstream attributes. (2026-05-11-identity-group-rebac)
- TypeScript 5.x / Node 20+ for the BFF; Python 3.11+ for Dynamic Agents + Next.js route handlers, existing BFF auth helpers, FastAPI, Starlette middleware, OpenFGA HTTP API, existing ReBAC tuple helpers (2026-05-16-dynamic-agent-pdp-gate)
- Existing OpenFGA store and MongoDB-backed Dynamic Agent records; no schema migration required (2026-05-16-dynamic-agent-pdp-gate)
- Python 3.13 runtime in the scanner image; TypeScript client behavior only for verification if needed +
cisco-ai-skill-scanner, FastAPI scanner API, Docker/Helm scanner packaging, Next.js server-side scanner client (2026-05-13-skill-scanner-load-error) - N/A — no persisted storage change (2026-05-13-skill-scanner-load-error)
- TypeScript 5.x / Node 20+ for Next.js 16 + React 19 UI/BFF; Python 3.11+ for Dynamic Agents and MCP auth paths; Go for the GitHub MCP server + Next.js App Router, NextAuth/OIDC session and JWT validation, MongoDB driver, OpenFGA HTTP API, Keycloak OIDC/Admin APIs, Dynamic Agents FastAPI services, existing MCP auth middleware, AWS KMS or equivalent KMS for production key wrapping (2026-05-20-openbao-credential-exchange)
- MongoDB credential metadata and encrypted payload collections; OpenFGA tuples for
secret_refand connector/provider use; KMS/CMK for wrapping per-credential data keys; no OpenBao datastore in the first release (2026-05-20-openbao-credential-exchange)
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.
- 1mo ago First seen · 64 lines · 1,906 tokens per session scan A 442ec2d80ec6
specify-rules is a cursor rule published in the GitHub repository caipe-io/ai-platform-engineering (432 stars, last pushed today), licensed Apache-2.0. It adds 1,906 tokens to every session, about $0.0076 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 2026-08-30.
Other cursor rules, from other repositories
heartbeat-execution-boundaries
Keep heartbeat a clear lifecycle coordinator with cohesive modules, explicit execution authority, and safe resource ownership.
kandev-pr-fixup
Kandev PR fixup workflow prompt for Cursor agents.
conventional-commits
Always use Conventional Commits for git commit messages.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.