Agent
Maintains IDE adapter definitions and registration flow. Use when adding or updating built-in adapters and their mapping behavior.
Agent
Maintains IDE adapter definitions and registration flow. Use when adding or updating built-in adapters and their mapping behavior.
Agent
Owns sync pipeline logic in core modules. Use when changing scanner, mapper, linker, reconciler, state, or incremental sync behavior.
Agent
Maintains documentation indexes and platform docs consistency. Use when updating docs structure, links, terminology, and cross-platform navigation.
Agent
Builds and maintains CLI end-to-end tests with real filesystem behavior. Use for sync, unlink, init, and integration regressions in test-e2e.
Agent
Performs final validation before merge or release. Use to audit completeness, run checks, and report residual risks without expanding scope.
Agent
You are a test agent used to verify local sync behavior.
Agent
Designs, writes, and updates automated tests for agentlink with the correct unit-vs-e2e boundary, repository conventions, and regression-focused assertions. Use when the user asks to add test coverage, fix failing tests, or strengthen test quality.
Agent
Writes and updates focused unit tests with MockFs and repository test conventions. Use for logic-level coverage in src//tests.
Command
Apply a safety protocol for changes under src/adapters/ and related mappings.
Command
Resolve pull request comments with traceable changes and clear closure notes.
Command
Run a focused code review for current changes with a risk-first mindset.
Command
Create a GitHub release with clear notes, traceable changes, and correct tag/version alignment.
Command
Prepare a high-quality pull request with clear intent, risk notes, and test plan.
Command
Audit documentation consistency against current implementation and workflows.
Command
Publish the package to npm using a safe, repeatable release sequence.
Command
Run a pre-release readiness check for quality, risk, and documentation completeness.
Command
Run the full test suite, diagnose failures, and implement minimal safe fixes.
Command
Validate agentlink sync behavior for one or more targets end-to-end.
Command
Run this command to validate command-file syncing.
Skill Claude CodeCodex
Creates portable agent definitions for OpenCode, Cursor, and Claude Code from one canonical .ai/agents/.md file. Use when the user asks to create a new agent or subagent, define agent behavior, or configure agent capabilities.
Skill Claude CodeCodex
Creates end-to-end tests for agentlink CLI commands using real filesystem behavior and repository conventions. Use when the user asks to add, update, or review tests in test-e2e, especially for sync, unlink, init, state persistence, and symlink lifecycle behavior.
Skill Claude CodeCodex
Creates portable Agent Skills that work across Cursor, Claude Code, OpenCode, and Windsurf from one canonical definition. Use when the user asks to create a new skill, design a reusable skill workflow, or make one skill compatible with multiple agentic IDEs.
Skill Claude CodeCodex
Creates and updates unit tests in src//tests using agentlink conventions, MockFs, and focused function-level scenarios. Use when the user asks to add unit coverage, validate pure logic, or test command/core/adapter behavior without real filesystem e2e setup.
Instructions file CodexOpenCode
AGENTS.md instructions for bohorquezod/agentlink, a project described as: Node.js CLI to generate granular symlinks from .ai to agent IDE configs.