Skills（SKILL.md）は、AIエージェント（Claude Code、Cursor、Codexなど）に特定の能力を追加するための設定ファイルです。
詳しく見る →Find similar vulnerabilities across a codebase after discovering one instance. Uses pattern matching, AST search, Semgrep/CodeQL queries, and manual tracing to propagate findings. Adapted from Trail of Bits. Use after finding a bug to check if the same pattern exists elsewhere.
>-
This skill helps you discover and install skills from the open agent skills ecosystem (https://skills.sh). Based on the [find-skills](https://github.com/vercel-labs/skills/tree/main/skills/find-skills
>
>
>
>
> **Nightly usage (2026-04-09):** `$dream start` now runs harvest as part
'Manage the PROGRAM.md operational contract that bounds autonomous development loops. Use when setting up or validating PROGRAM.md/AUTODEV.md, choosing between autodev and evolve, or turning a repeated postmortem -> analyze -> plan -> pre-mortem -> implement -> validate cycle into the v2 autonomous workflow.'
Monitor the knowledge flywheel health.
>
Track progress through the RPI workflow with permanent gates.
Compute metrics for Claude Code sessions. Discovers via ccrider, filters trivial, computes friction/opportunity/fingerprint scores. Use for broad session triage.
Output rules for all agents - concise, scannable, actionable. Based on Matt Pocock's planning principles.
Core principle that the main agent is a coordinator, not an implementer. All work must be delegated to subagents.
Produce an implementation plan at `.turbo/plans/<slug>.md`. Capture the task, survey patterns, escalate decisions, discuss, and draft.
Run two independent interpretations of third-party feedback in parallel (internal + codex peer), then reconcile into enriched items with clear intent summaries. Designed for feedback where the author'
Provides Python patterns for type-first development with dataclasses, discriminated unions, NewType, and Protocol. Must use when reading or writing Python files.
yolo
Compile all gathered knowledge into actionable intelligence. Validate hypotheses against evidence. Produce structured reports with traceable findings.
Criterios de evaluación para auditorías de funnel NGX, incluyendo checklist visual, UX, copy y conversión
Create and validate YAML diagram files. Use when writing new diagrams or troubleshooting YAML syntax.
This skill enables control of macOS applications and system functions through AppleScript execution via the `osascript` command.
Cloud infrastructure design and deployment patterns for AWS, Azure, and
Development standards for the Nathan n8n-Jira agent automation system.
Security architecture and implementation patterns. Use when designing
Map skills to proficiency levels using CEFR, Bloom's taxonomy, and DigComp frameworks. Use when designing skill progressions or assessing learner proficiency.
Create a thoughtful, supportive commit message from your staged changes.
>
>
>
Senior Quality Manager Responsible Person (QMR) for HealthTech and MedTech companies. Provides overall quality system responsibility, regulatory compliance oversight, management accountability, and strategic quality leadership. Use for quality system governance, regulatory compliance oversight, management responsibility, and quality strategic planning.
>
>
When the user asks about roadmap planning, sprint planning, milestone estimation, or capacity planning
Hooks are event-driven automation scripts that execute in response to Claude Code events. Use hooks to validate operations, enforce policies, add context, and integrate external tools into workflows.
Add a new third-party dependency to the project following the version catalog and approval workflow.
>
Transfer knowledge from source domains to novel target situations through structural mapping. Use when facing new markets, products, or situations where past experience provides relevant patterns. Produces adapted solutions with explicit mappings and context adjustments.
Execute evidence-based decision-making through 6-stage causal flow. Use for known processes, operational execution, and decisions with clear cause-effect chains.
Evaluate alternative scenarios by simulating interventions on past decisions or hypothetical futures. Use when assessing decisions in hindsight, planning scenarios, or comparing paths not taken. Produces comparative analysis with probability-weighted outcomes.
SOLID object-oriented design principles for maintainable code
Use when creating new skills, documenting workflows, or standardizing processes into reusable skill definitions.
Use when starting, pausing, completing, or transitioning task status in the development workflow.
Use when making technical decisions, choosing technologies, or documenting architectural choices. Creates ADRs (Architecture Decision Records).
Context optimization extends the effective capacity of limited context windows through strategic compression, masking, caching, and partitioning. The goal is not to magically increase context windows
日常科研 - Stella建立临时研究站,进行系统的科学观察和实验,享受科学家的日常
When Truth Layer identifies a blocker:
When you discover a blocker:
Coordinate multi-section document creation by delegating to doc writer agents through VibeKanban task management. The coordinator (Chief of Staff) plans and monitors but never writes content directly.