AmadeusITGroup

Write a clear Conventional Commits message from a set of staged changes. Triggers: user says 'write a commit message', 'draft a commit', or pastes a diff and asks how to describe it. Use whenever changes are staged and need a well-formed commit subject and body.

Turn a set of merged pull requests into a clean, grouped release-notes section. Triggers: user says 'write release notes', 'summarize this release', 'changelog for these PRs'.

Generate evaluation test cases for any skill by browsing its SKILL.md and references. Produces evals.json with [auto]-prefixed expectations. Triggers: user says 'generate evals', 'create evals for skill', 'scaffold evals'; or after creating/updating a skill that needs test coverage. Use this skill whenever a skill exists but has no evals, or when evals need updating after skill changes.

Run skill evaluations by spawning sub-agents for test execution and LLM-as-judge grading. Includes blind A/B comparison and eval-viewer for visual review. Triggers: user says 'run evals', 'evaluate skill', 'test skill'; or after generating/updating a skill with evals.

Parallel codebase assessment with specialized subagents for QA, Technical Debt, Functional, and Architecture analysis. Use when analyzing a whole codebase for quality, performing code audits, assessing code health, identifying improvements, evaluating technical debt across a project, or when onboarding to a new codebase. Even if the user just says "review this project" or "what should we improve", this skill applies.

Parallel PR review with specialized subagents for QA, Technical Debt, Functional, and Architecture analysis. Use when reviewing pull requests, checking test quality, detecting tech debt, validating functional completeness, or verifying architectural consistency.

Triage a repository's open issue backlog by proposing priorities, labels, and comments. Detects already-triaged issues, identifies potential duplicates, and generates a gh-based apply script.

Triage a repository's open pull request backlog by proposing priorities, labels, and comments. Detects already-triaged PRs, identifies potential duplicates, and generates a gh-based apply script.

This skill should be used when the user asks to "add a rule", "create a ruleset", "add a fact", "create a custom operator", "add a rules-engine action", "make behavior conditional at runtime", or when authoring or debugging anything in an Otter `@o3r/rules-engine` workspace. Guides the fact/operator/action/ruleset workflow, registration steps, and safe defaults.

Use and create @ama-sdk plugins for request transformation, authentication, resilience, and response processing. Use when user asks about SDK plugins, wants to add auth/retry/timeout, needs to create a custom plugin, or wants to transform requests before sending, wrap the fetch call itself (logging, metrics, circuit-breaking), or process/transform responses.

This skill should be used when the user asks to "create an E2E test", "write a Playwright scenario", "add a Playwright test", "generate e2e tests", "create actions and checks", "write playwright checks", "add scenario for feature", or when modifying, reviewing, or debugging Playwright E2E test files in the workspace.

This skill should be used when the `@o3r/mcp` MCP server is not available and needs to be installed, or when the user asks to "install otter mcp", "setup otter mcp", or "configure otter mcp server".

This skill should be used when the user asks to "add configuration to a component", "introduce a config flag", "add a feature toggle", "add a runtime config", "create a config property", or when any new configuration flag is being introduced in an Otter workspace. Guides scope decisions, safe defaults, and absence-detection patterns.

This skill should be used when the user asks to "create a component", "generate a library", "scaffold a service", "add a fixture", or when creating any Angular artifact in an Otter workspace. Ensures schematics are used over manual file creation.

Answer questions about using AI Primitives Hub and authoring bundles. Use this skill whenever the user asks about bundles, collections, hubs, profiles, sources, scopes, marketplace, repository installation, deployment manifests, source scaffolding, collection schemas, validation errors, publishing, or any AI Primitives Hub extension feature — even if they don't name the extension explicitly.

Performs thorough code reviews following best practices. Use when reviewing pull requests, checking code quality, or ensuring coding standards compliance.