Create or update root and nested AGENTS.md files that document scoped conventions, monorepo module maps, cross-domain workflows, and (optionally) per-module feature maps (feature -> paths,...

Change persistent state with checkpoint and rollback support. Use when modifying files, updating databases, changing configuration, or any operation that permanently alters state.

Request clarification when input is ambiguous. Use when user request has missing parameters, conflicting interpretations, or insufficient constraints for reliable execution.

Analyzes conversations to extract patterns and workflows, then generates Skill.md files matching user's style. Use to convert successful chats into reusable skills.

Enforce policies, guardrails, and permission boundaries; refuse unsafe actions and apply least privilege. Use when evaluating actions against policies, checking permissions, or reducing scope to...

Merge outputs from multiple sources, resolve conflicts, and reconcile constraints into a unified result. Use when combining parallel agent outputs, merging data from different systems, or...

Convert data between formats, schemas, or representations with explicit loss accounting and validation. Use when reformatting data, mapping between schemas, normalizing inputs, or translating structures.

Implement code from a GitHub issue specification. Explores codebase, researches docs, creates temporary implementation plan, then implements incrementally (small piece → test → commit → next)....

Safely undo changes by restoring to a checkpoint. Use when verify fails, errors occur, or explicit undo is requested. Essential for the CAVR pattern recovery.

Create a safety checkpoint marker before mutation or execution steps. Use when about to modify files, execute plans, or perform any irreversible action. Essential for the CAVR pattern.

Retrieve prior decisions, rationale, and learned patterns from memory to apply consistently. Use when needing context from previous interactions, looking up past decisions, or ensuring consistency...

Break a goal into subgoals, constraints, and acceptance criteria. Use when planning complex work, creating work breakdown structures, or defining requirements.

Split work across subagents with explicit contracts, interfaces, and merge strategies. Use when parallelizing tasks, distributing workload, or orchestrating multi-agent workflows.

Ingest and parse incoming messages, events, or signals into structured form. Use when processing external inputs, handling API responses, parsing webhook payloads, or ingesting sensor data.

Produce a comprehensive audit trail of actions, tools used, changes made, and decision rationale. Use when recording compliance evidence, tracking changes, or documenting decision lineage.

Create a new artifact (text, code, plan, data) under specified constraints. Use when producing content, writing code, designing solutions, or synthesizing outputs.

Write stable learnings, decisions, and patterns to durable storage like CLAUDE.md or knowledge files. Use when saving project decisions, recording patterns, or updating long-term memory.

Emit a message or event to an external system with policy enforcement and approval gates. Use when publishing messages, calling APIs, sending notifications, or triggering external workflows....

Find failure modes, edge cases, ambiguities, and exploit paths in plans, code, or designs. Use when reviewing proposals, auditing security, stress-testing logic, or validating assumptions.

Implement secure coding practices following OWASP Top 10. Use when preventing security vulnerabilities, implementing authentication, securing APIs, or conducting security reviews. Triggers on...