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 rules/rajudandigam/agent-inspect/cursorrulesgit clone --depth 1 https://github.com/rajudandigam/agent-inspectWhat 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 | $0.00268 | $0.00268 |
| Opus 5 | $0.00134 | $0.00134 |
| Sonnet 5 | $0.00054 | $0.00054 |
| Haiku 4.5 | $0.00027 | $0.00027 |
Grade A, and why
cursorrules 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.
What it actually says
agent-inspect — Cursor guardrails
Source of truth
- Product, architecture, MVP scope, public API, data model, JSONL schema: repo top-level docs (start with
README.md) - Step-by-step implementation sequence: follow docs/implementation/CURSOR-MAINTAINER-ROADMAP.md for maintainer release trains; otherwise existing patterns in
packages/
MVP drift (do not implement unless the PRD explicitly promotes it)
- No SQLite
- No replay
- No token counting
- No cost calculation
- No framework adapters
- No browser dashboard
- No OpenTelemetry
- No plugin system
- No general-purpose logger API
agent-inspect must not become a logger library. The core concept is an execution tree of steps.
Engineering requirements
- TypeScript: strict mode; match repo
tsconfigsettings. - Tests: add or update tests when implementation behavior changes.
- Instrumentation safety: tracing, stepping, or observation must never throw into user code in a way that breaks the agent; failures degrade gracefully (see PRD).
General guidance
Please ask questions and get clarify if you are not clear on any decisions or implementations.
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 · 30 lines · 268 tokens per session scan A e2b2c59626c3
cursorrules is a cursor rule published in the GitHub repository rajudandigam/agent-inspect (406 stars, last pushed 5d ago), licensed MIT. It adds 268 tokens to every session, about $0.0013 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 cursor rules, from other repositories
changelog-posts
How we do changelog posts and format them.
state-management
Use the following stack. Do not introduce or recommend Redux or React Context for shared/global state.
ipynb-files
This file contains rules and prompts for handling Jupyter Notebook (.ipynb) files in Cursor IDE, which currently doesn't have native support for .ipynb files.
cursor
You are working on the checkout service. Preserve transaction integrity and auditability.
ipc
You're building an Electron app following good security practices.
graftcode-typescript-node-nextjs
Mandatory Graftcode architecture rules for TypeScript / Node.js / Next.js — expose plain modules via Graftcode instead of REST/gRPC/Thrift.