breaking: enhance agent creation and skill management#13
Merged
diegoitaliait merged 84 commits intomainfrom Mar 31, 2026
Merged
breaking: enhance agent creation and skill management#13diegoitaliait merged 84 commits intomainfrom
diegoitaliait merged 84 commits intomainfrom
Conversation
- Introduced a new markdown file for the agent creation system prompt detailing the process for generating agent configurations. - Created a comprehensive guide on system prompt design patterns to aid in writing effective agent prompts. - Added best practices for triggering examples in agent descriptions to ensure reliable agent activation. - Implemented a validation script for agent markdown files to check for required fields, structure, and content quality.
…and testing skills; update AGENTS.md to reflect current skill structure.
- Implemented `improve_description.py` to enhance skill descriptions based on evaluation results using Claude. - Created `package_skill.py` for packaging skill folders into distributable .skill files. - Added `quick_validate.py` for basic validation of skill structure and frontmatter. - Developed `run_eval.py` to evaluate skill descriptions against a set of queries and output results. - Introduced `run_loop.py` to automate the evaluation and improvement loop for skill descriptions. - Added utility functions in `utils.py` for parsing SKILL.md files.
- Introduced internal skills for analyzing change impacts, generating PR descriptions, and creating/modifying project components in Java, Node.js, Python, and Bash. - Added a skill for syncing global Copilot configurations into repositories with a three-phase sync model. - Implemented a Terraform skill for managing resources and modules with a decision guide for feature vs module usage. - Updated AGENTS.md to reflect the new internal skills and their paths.
- Implemented `package_skill.py` to create distributable .skill files from skill folders, including validation checks for SKILL.md and exclusion of unnecessary files. - Created `quick_validate.py` for basic validation of skill frontmatter in SKILL.md, ensuring required fields and allowed properties are present. - Introduced `run_eval.py` to evaluate skill descriptions against a set of queries, determining if the skill triggers correctly and outputting results in JSON format. - Developed `run_loop.py` to facilitate an iterative process of evaluating and improving skill descriptions until all queries pass or a maximum number of iterations is reached. - Added utility functions in `utils.py` for parsing SKILL.md files, extracting name and description from the frontmatter.
…nce and naming conventions
…on, Terraform, Planner, and PR Editor
…tion - Deleted the validate-copilot-customizations.sh script from the repository. - Updated AGENTS.md to remove references to the deleted validation script. - Modified CONTRIBUTING.md to reflect changes in validation requirements, replacing mentions of the script with make commands. - Adjusted COPILOT_REVIEW.md to align validation steps with the current toolchain, removing references to the script. - Updated Makefile to remove the validator variable and adjust linting commands to handle the absence of scripts gracefully.
…s and clarify approved upstream scope
…pository detection and codebase scanning rules
…tailed sync procedures and naming conventions
…d guidance on asset preservation and sync procedures
…d instruction asset guidelines
…heir usage guidelines
…consolidate and rename skills for clarity and consistency
- Introduced a new script `.github/scripts/validate-copilot-customizations.sh` to validate core Copilot customization invariants. - Updated `.github/security-baseline.md` to reference the new validation script. - Removed obsolete skills: `awesome-copilot-suggest-awesome-github-copilot-agents`, `awesome-copilot-suggest-awesome-github-copilot-instructions`, and `awesome-copilot-suggest-awesome-github-copilot-skills`. - Updated `AGENTS.md` to reflect the removal of obsolete skills. - Modified `CONTRIBUTING.md` to clarify naming conventions for assets. - Updated `COPILOT_REVIEW.md` to reflect changes in script names and improve clarity on deprecations. - Adjusted paths in `tests/test_contract_runner.py` to align with the new script naming conventions. - Updated the `internal-sync-global-copilot-configs-into-repo` skill to reflect changes in script and manifest paths.
- Created output templates reference for various formats in antigravity-web-scraper. - Added CHANGELOG.md for youtube-summarizer with version history and updates. - Developed SKILL.md for youtube-summarizer detailing purpose, usage, and workflow. - Implemented extract-transcript.py script for extracting YouTube video transcripts. - Created install-dependencies.sh script for installing required Python dependencies. - Updated AGENTS.md to include new skills and their documentation paths.
- Updated descriptions for antigravity-steve-jobs and antigravity-web-scraper skills for clarity and consistency. - Introduced new skills: awesome-copilot-codeql, awesome-copilot-dependabot, awesome-copilot-postgresql-optimization, awesome-copilot-secret-scanning, awesome-copilot-sql-optimization, internal-changelog-automation, internal-devops-core-principles, internal-kubernetes-deployment, internal-performance-optimization, and internal-skill-management to improve repository governance and operational efficiency. - Enhanced internal documentation for editing presentations and skill management processes. - Refined agent synchronization and skill consolidation strategies in AGENTS.md.
- Deleted obsolete `utils.py` script from skill-creator. - Added new skills: `internal-agent-authoring`, `internal-agents-md-bridge`, `internal-copilot-audit`. - Updated `internal-skill-management` to improve descriptions and authoring workflow. - Enhanced `internal-sync-global-copilot-configs-into-repo` with clearer guidance and updated skill references. - Revised `AGENTS.md` to include new agents and remove deprecated ones. - Created `OPTIMIZATION_PLAN.md` to outline remaining optimization tasks and completed items. - Added tests for the validation script to ensure functionality and correctness.
- Deleted `redlining.py` and `thumbnail.py` scripts as they are no longer needed. - Added new skill `internal-agent-development` for repository-owned agent creation and management. - Updated `internal-sync-global-copilot-configs-into-repo` skill to reference `internal-agent-development`. - Enhanced documentation for `obra-dispatching-parallel-agents`, `obra-remembering-conversations`, `obra-subagent-driven-development`, and `obra-using-skills` with compatibility notes for Claude Code. - Revised `obra-pulling-updates-from-skills-repository` skill description to reflect upstream changes. - Introduced `obra-writing-skills` skill for authoring and validating skills. - Updated `AGENTS.md` to reflect changes in skill names and descriptions. - Cleaned up `OPTIMIZATION_PLAN.md` to remove completed tasks and adjust skill counts.
- Introduced `internal-principal-cloud-aws`, `internal-principal-cloud-azure`, and `internal-principal-cloud-gcp` agents for principal-level architecture review and tactical execution planning in AWS, Azure, and GCP respectively. - Added `internal-quality-engineering` agent for test strategy, performance diagnosis, and observability design. - Created `internal-sync-control-center` agent for managing Copilot customization assets and repository governance. - Implemented `internal-skill-development` and `internal-agent-development` skills for creating and refining repository-owned skills and agents. - Updated `AGENTS.md` to reflect new agent names and roles, consolidating governance responsibilities. - Removed deprecated `obra-writing-skills` skill and updated references to align with new internal skill guidelines. - Adjusted various scripts and skills to ensure proper integration and functionality within the repository.
…te AGENTS.md with new naming conventions and remove deprecated agent
…enhance instruction handling and skill declaration checks
- Deleted outdated Terraform Azure and AWS specific instructions to streamline governance rules. - Removed general Terraform instructions to avoid redundancy and focus on Azure-specific best practices. - Updated internal Docker instructions to enhance security and reproducibility standards. - Improved internal GitHub Actions instructions for better security and performance. - Enhanced internal Terraform instructions with additional documentation and testing guidelines. - Cleaned up AGENTS.md to reflect the removal of obsolete instruction files.
…cument evidence, implications, and actionable insights
…actions, Lambda, Python, Docker, and Terraform; improve clarity and usability
…in consistent coding styles and versioning
…ma and JSON/Markdown outputs
…tional Skills" - Updated all internal agent markdown files to change the section heading from "Declared Skills" to "Preferred/Optional Skills" to better reflect the nature of the skills listed. - Modified the skill usage contract in each agent to clarify that preferred or optional skills should be treated as a balanced set of options, without default prioritization of internal skills over imported ones. - Adjusted validation scripts and tests to accommodate the new skill section naming and ensure proper functionality. - Revised documentation and templates to align with the new skill section structure, emphasizing the optional nature of skill guidance.
…transcript-api - Created `extract-transcript.py` to extract YouTube video transcripts with options for language and listing available transcripts. - Added `install-dependencies.sh` to automate the installation of the `youtube-transcript-api` dependency.
- Deleted the following skills due to redundancy and lack of focus: - awesome-copilot-architecture-blueprint-generator - awesome-copilot-create-github-action-workflow-specification - awesome-copilot-create-implementation-plan - awesome-copilot-create-readme - awesome-copilot-documentation-writer - awesome-copilot-refactor-plan - Updated AGENTS.md to reflect the removal of these skills and adjusted the skill count in zOptimizer-final.md to maintain accurate documentation.
…avity-code-review-checklist`, consolidate into `internal-script-bash` and `internal-code-review`. Update routing rules and documentation for improved clarity and structure. Enhance Bash scripting guidelines with defensive patterns and cleanup practices. Adjust agent references and skill counts in documentation to reflect changes.
…hon skills; updated internal project descriptions and best practices. Consolidated testing guidance and improved clarity on project structures across multiple languages. Enhanced documentation for internal project skills to reflect modern practices and frameworks.
…r clarity and accuracy
… zOptimizer-final.md for clarity and focus
…hy, consolidating skills, and updating internal documentation for clarity and focus
… structures, and enhance documentation - Deleted several obsolete skills: awesome-copilot-create-github-pull-request-from-specification, awesome-copilot-instructions-blueprint-generator, awesome-copilot-postgresql-optimization, awesome-copilot-sql-optimization, and terraform-terraform-style-guide. - Updated internal-agent-development and internal-performance-optimization skills to improve clarity and governance. - Enhanced internal-pr-editor to support specification-aware drafting. - Adjusted internal-sync-global-copilot-configs-into-repo to reflect changes in guidance refresh order. - Cleaned up AGENTS.md to remove references to deleted skills and ensure accurate documentation. - Finalized cleanup of deprecated frontmatter in imported agents.
…ical Thinking, and DevOps Expert agents for enhanced functionality
…nts and skills to enhance clarity and remove deprecated elements
…aration in internal agents - Updated references from `validate-copilot-customizations.sh` to `validate-copilot-customizations.py` across various files, including PR templates, agent definitions, and documentation. - Enforced the requirement for internal agents to explicitly declare `tools:` in their frontmatter to enhance clarity and maintainability. - Added validation checks to ensure that internal agents without a `tools:` declaration are flagged during the validation process. - Adjusted tests to reflect the new validation rules and ensure compliance with the updated agent structure.
…ons.sh` to `validate-copilot-customizations.py` and update usage instructions
…documentation for consistency
…iles for clarity and consistency
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Introduce a comprehensive agent creation system with validation scripts, improve skill descriptions, and remove obsolete documentation. Update internal governance and enhance documentation for clarity and consistency across various skills and agents.
Change Type
List of changes
Consumer Impact
Testing Instructions
Validation Evidence
Breaking Changes
Checklist
validate-copilot-customizations.sh --mode strictexecuted successfully