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/eliornl/rolemule/cursorrulesgit clone --depth 1 https://github.com/eliornl/rolemuleWhat 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 | $0.07759 | $0.07759 |
| Opus 5 | $0.03880 | $0.03880 |
| Sonnet 5 | $0.01552 | $0.01552 |
| Haiku 4.5 | $0.00776 | $0.00776 |
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 yesterday.
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 — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RoleMule — Cursor Rules Index
Product name: RoleMule (repo folder: rolemule).
All detailed coding rules live in .cursor/rules/. This file is the index.
Rule Files
| File | Covers |
|---|---|
rolemule-core.mdc |
App name, error system (APIError/ErrorCode, CFG_6xxx no-API-key, RES_3002 duplicate workflow, post-start duplicate / uq_user_job_company, content fingerprint NFKC + utils/application_dedupe.py), POST /workflow/start job file uploads (.pdf/.txt/.docx, 5 MB, _JOB_FILE_MAGIC), background task DB, workflow failure (no partial outputs, applications list join) |
python-conventions.mdc |
Python style, file sections, FastAPI endpoint template |
database-patterns.mdc |
SQLAlchemy JSONB, flag_modified(), soft delete, partial indexes, race conditions, migrations, UserWorkflowPreferences upsert |
auth-patterns.mdc |
JWT structure, _make_jwt, all auth endpoints, token revocation, lockout, timing attack fix |
security-python.mdc |
XSS, file upload MIME, secrets vs random, no-JWT-in-URL rule, user enumeration prevention |
codeql-security-scanning.mdc |
CodeQL CI, sanitize_log_value() / mask_email() logging, query-filters, secret scanning test keys, Pydantic field_validator, extension entity decode |
security-middleware.mdc |
Middleware stack, CORS explicit headers, no wildcard origins, maintenance mode, CSP nonce-based style-src (no 'unsafe-inline'), .is-hidden / CSS utility class reference |
settings-and-env.mdc |
All env vars, get_settings(), ENCRYPTION_KEY requirement, .env commit rules |
llm-integration.mdc |
Multi-provider LLM (get_llm_client), per-user BYOK (Gemini/OpenAI/Anthropic/Ollama), model allowlists in utils/llm/models.py, grounding, CFG_6001, asyncio.wait_for(), generate_stream / SpeakFieldStreamer, JSON parsing |
agent-patterns.mdc |
5 workflow agents + standalone on-demand (interview prep, CV optimizer, Practice Interview, Hiring Outreach) + 6 career tools, any workflow agent failure fails the run, workflow_preferences injection, BYOK model override, Company Research — _has_usable_company_name / unnamed-posting / disambiguation / staffing / research_quality |
interview-prep-feature.mdc |
Standalone agent, background task, Redis atomic lock, rate limit (static Interview tab — not Practice Interview) |
cv-optimizer-feature.mdc |
CV Optimizer loop — hiring manager + reviser agents, cv_optimization JSONB, ownership-before-cache, WebSocket iteration events, Optimize CV tab |
mock-interview-feature.mdc |
Practice Interview — HR/Pro/Manager, duration minutes, browser STT/TTS, BYOK, speak streaming / speak_delta, ws-guard, mock_interview JSONB |
hiring-outreach-feature.mdc |
Hiring Outreach — public web contact finder + copy-only draft messages, hiring_outreach JSONB, Redis lock, WS hiring_outreach_*, 10th Outreach tab |
career-tools.mdc |
6 tools, endpoints, rate limits, agent implementation pattern, output JSON schemas, copy button pattern, no-placeholder rules, UI CSS classes |
caching-redis.mdc |
Cache TTLs, helpers, rate limiting, auth-specific Redis keys |
websocket-patterns.mdc |
Endpoints, connection limits, broadcast helpers, WS-only JWT query-param exception, mock interview speak_delta + client session filter |
logging-patterns.mdc |
StructuredLogger, sanitize_log_value() / mask_email(), request context, sensitive data redaction, exc_info=True requirement, bulk-script safety |
google-oauth.mdc |
3-step OAuth flow, CSRF state in Redis, exchange-code pattern, open-redirect prevention |
email-and-misc-utils.mdc |
Gmail SMTP, resume parser, BYOK encryption |
frontend-js-strict.mdc |
TypeScript strict mode, null safety, anti-patterns table, event bus, no style= HTML attributes |
landing-page.mdc |
Landing page sections, screenshot showcase (.ss-tabs, .ss-panel, ssActivateTab), screenshot file locations, update workflow |
dashboard-home.mdc |
Dashboard application list — list/stats join workflow_sessions (hide workflow-failed), funnel stats (Applied card + response rate formula), search/filter/sort, toasts (notifyReady c:/f: dedupe keys, formatWorkflowFailureDetail, duplicate headline), isPlaceholderCompanyName / displayCompanyNameOrUnknown (Unknown employer label), card CSS, auth guard, EXISTS pagination / Load More dedupe |
ui-application-detail.mdc |
10-tab layout (Optimize CV + Practice Interview + Outreach), sub-tab patterns (Resume + Interview), cover letter card, ui/src/pages/cv-optimizer.ts, ui/src/pages/mock-interview.ts, ui/src/pages/hiring-outreach.ts, View posting / additional_locations, render functions, isPlaceholderCompanyName / Unknown employer / About this opportunity, company research notices (research_quality / staffing), all CSS classes |
accessibility.mdc |
WCAG 2.1 AA, heading hierarchy, landmarks, aria-label rules |
analytics-consent-onboarding.mdc |
PostHog, cookie consent API, onboarding tour |
chrome-extension.mdc |
Extension structure, popup vs service-worker submit paths, source_url → job_url, dev/prod IS_DEV URL toggle, popup design system, icon generation, landing-page mockup sync, extractPageContent split-view / detail-pane root (popup.js + service-worker.js), form autofill, InputMethod.EXTENSION |
adding-new-features.mdc |
Checklists: new endpoint, new agent, new tool, new migration, new static asset, new preference |
settings-page.mdc |
Settings tabs, Preferences tab, AI Setup tab, auto-save pattern, custom slider/toggle/dropdown, account-icon color variants, self-hosted Font Awesome |
frontend-build-pipeline.mdc |
Vite+TS page entries + esbuild CSS, asset_url(), manifest, CI/Docker, make build-frontend |
cli.mdc |
RoleMule CLI — Typer commands, rolemule_client, tests, docs |
mobile-responsive.mdc |
Breakpoints, navbar collapse (navbar-expand-xl landing / navbar-expand-lg dashboard), scrollable tab bars, app.css global mobile utilities, per-page patterns |
unit-testing.mdc |
~300 agent unit tests + ~1,050 API integration tests + 364 CLI mocked tests + 6 ASGI CLI integration tests; live_server_helpers; ASGITransport (not TestClient) for CLI ASGI; cov ≥97% includes root test_cv_*; NullPool; asyncio_default_fixture_loop_scope=session |
e2e-testing.mdc |
~1,380 Playwright tests; CI runs smoke only (e2e-smoke / @smoke); full suite is local/optional; setupAuth + page.route() patterns; JWT/cookie-consent setup |
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.
- yesterday First seen · 132 lines · 7,759 tokens per session scan A b62c45f35134
cursorrules is a cursor rule published in the GitHub repository eliornl/rolemule (37 stars, last pushed 2d ago), licensed MIT. It adds 7,759 tokens to every session, about $0.0388 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
cursorrules
Agent2 is a framework for building production AI agents with typed HTTP APIs. PydanticAI handles the agent loop. Agent2 handles everything else: API, auth, pause/resume, approvals, provider routing, knowledge search.
product-manager
Holistic product leader who owns the full product lifecycle — from discovery and strategy through roadmap, stakeholder alignment, go-to-market, and outcome measurement. Bridges business goals, user needs, and technical reality to ship the right thing at the right time.
api-tester
Expert API testing specialist focused on comprehensive API validation, performance testing, and quality assurance across all systems and third-party integrations.
developer-advocate
Expert developer advocate specializing in building developer communities, creating compelling technical content, optimizing developer experience (DX), and driving platform adoption through authentic engineering engagement. Bridges product and engineering teams with external developers.
architectural-and-structural-rules
The project follows a layered architecture with clear separation of concerns.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.