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/sruja-ai/sruja/cursorrulesgit clone --depth 1 https://github.com/sruja-ai/srujaWhat 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 | $0.00283 | $0.00283 |
| Opus 5 | $0.00142 | $0.00142 |
| Sonnet 5 | $0.00057 | $0.00057 |
| Haiku 4.5 | $0.00028 | $0.00028 |
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
Sruja Architecture Context
This file is auto-generated by sruja sync-ide-rules -r .
Architecture Overview
- Crates: 15
- Databases: 1
Layers
CORE ENGINE (7 crates)
No external dependencies allowed.
EXTRACTION (2 crates)
No external dependencies allowed.
DELIVERY (2 crates)
No external dependencies allowed.
SECONDARY (4 crates)
No external dependencies allowed.
Forbidden Patterns
- Lower-tier crates must not depend on higher-tier crates
- sruja-cli is the top-level aggregator — no other crate should depend on it
- WASM-only crates must not use native-only APIs (tree-sitter, fastembed)
Architecture retrieval (MCP)
MCP: sruja_list_architecture_index → sruja_get_topology → sruja_get_elements → sruja_get_task_context (cache_friendly optional).
Compact brief: llms-architecture.txt. Full DSL only via MCP sruja://architecture/main.
When suggesting code
- Respect layer boundaries - check imports before suggesting
- Use existing patterns in the codebase
- If adding a new dependency, verify it does not violate boundaries
- Run
sruja drift -r .after changes to verify architecture health
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 · 41 lines · 283 tokens per session scan A d4ec42200633
cursorrules is a cursor rule published in the GitHub repository sruja-ai/sruja (22 stars, last pushed 9d ago), licensed Apache-2.0. It adds 283 tokens to every session, about $0.0014 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
boffin-pack-routing
ParselFire Core pack activation and stage pipeline.
boffin-post-change-audit
Pre-flight review and post-change kernel audit.
boffin-cpp-routing
Activate ParselFire Core C++ pack index.
dotnet-clean-code
C#/.NET-specific clean code standards and idioms. Covers naming conventions, C# language patterns, async/await correctness, DI, and .NET tooling hygiene. Complements engineering-principles.mdc.
api-design
API design guidelines — REST principles, GraphQL patterns, versioning, pagination, error handling, documentation, security, caching, and testing. Applied when designing or reviewing API endpoints.
engineering-security
Universal security principles — stack-agnostic. Applied when writing any code that handles input, authentication, authorisation, secrets, or sensitive data. Aligned with OWASP Top 10.