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/agent-patternsgit 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 | $0.00000 | $0.03062 |
| Opus 5 | $0.00000 | $0.01531 |
| Sonnet 5 | $0.00000 | $0.00612 |
| Haiku 4.5 | $0.00000 | $0.00306 |
Grade A, and why
agent-patterns 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 — 256 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Patterns
Two Types of Agents
| Type | Examples | Invocation |
|---|---|---|
| Workflow agents | job_analyzer, profile_matching, company_research, cover_letter_writer, resume_advisor |
Orchestrated by LangGraph in job_application_workflow.py |
| Standalone agents | interview_prep, mock_interview, hiring_outreach, hiring_manager + cv_optimizer_loop (orchestrator), all 6 career tools |
Called directly from API endpoint or background task |
Never add a standalone agent to the LangGraph graph — they run independently.
Workflow Execution Order (5 agents)
Step 1 (sequential): Job Analyzer
↓
Step 2 (sequential): Profile Matcher ← GATE DECISION
│ score >= threshold → continue automatically
└ score < threshold → AWAITING_CONFIRMATION (user must confirm)
↓
Step 3 (sequential): Company Research
↓
Step 4 (PARALLEL): Cover Letter Writer + Resume Advisor
Profile Matcher produces the Strategy tab content — it outputs competitive_positioning and application_strategy fields in addition to the match score and fit breakdown. There is no separate "Strategy Advisor" agent.
Cover Letter Writer and Resume Advisor run in parallel — they do not depend on each other; do not introduce a sequential dependency between them.
Job Analyzer — multiple locations
agents/job_analyzer.py extracts additional_locations (List[str]) for every office listed beyond the primary job_city / job_state / job_country. Prompt rule: capture all listed locations — never drop extras. Stored on JobAnalysisResult / workflow_sessions.job_analysis. Profile Matcher considers all offered locations for location-fit scoring.
Company Research — missing or placeholder employer name
agents/company_research.py uses _has_usable_company_name(name) before treating company_name as a real employer. It returns false for empty/whitespace, common LLM text placeholders (null, unknown, n/a, …), and dash-only strings (regex: hyphen / en dash / em dash / Unicode minus, with optional whitespace). When false, research runs in unnamed-posting mode using _format_job_context_for_unnamed_employer() and job-context prompts — the workflow does not fail solely for a missing employer. Frontend isPlaceholderCompanyName() mirrors this intent for display (see dashboard-home.mdc / ui-application-detail.mdc).
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 · 256 lines · 0 tokens per session scan A c7b6c59d4fa0
agent-patterns is a cursor rule published in the GitHub repository eliornl/rolemule (37 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,062 tokens. 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.