3 commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Darren Hinde
|
9625e43d92
|
feat: OpenCode-compliant agents with centralized metadata system (#151)
* chore: bump version to v0.8.0
* chore: update version to 0.7.1 instead of 0.8.0
* docs: expand comparison table with Oh My OpenCode and additional features
* backup: development context before reorganization
* refactor(context): reorganize development folder
- Remove 7 duplicate files (ui/design files exist in ui/web/)
- Remove 2 redundant READMEs (use navigation.md instead)
- Move react-patterns.md to frontend/react/
- Create frontend/react/navigation.md
- Update 4 navigation files (development, ui, frontend, frontend/react)
- Update developer profile (4 path changes: ui-styling-standards, design-systems, design-assets, animation-patterns)
- Update 3 core context files (visual-development.md, design-iteration.md, file-locations.md)
Result: 29 context files (down from 36), no duplicates, clean structure
Refs: REORGANIZATION_PLAN.md, REORGANIZATION_SUMMARY.md
* docs(context): add frontend delegation guide
- Create when-to-delegate.md with clear decision criteria
- Decision matrix for when to delegate vs. handle directly
- Common patterns and examples
- Red flags and green flags
- Frontend-specialist capabilities reference
- Update frontend and development navigation files
Helps orchestrators decide when to delegate UI work to frontend-specialist subagent.
* feat: enhance developer and advanced profiles with UI, development, and system builder context
- Developer profile: Add UI/UX patterns, development principles, design systems, React patterns, animation patterns, styling standards, and design assets
- Full profile: Add same UI/development context as developer profile for consistency
- Advanced profile: Add comprehensive system builder context including templates, guides, context system standards, and repository management context
- Update profile descriptions to reflect new capabilities
- All context files verified to exist and are properly referenced
* fix: remove duplicate context entries with incorrect paths
- Remove duplicate design-assets, ui-styling-standards, clean-code, animation-patterns, design-systems, react-patterns, and api-design entries that pointed to wrong paths
- Remove non-existent development-readme and mastra-ai-readme entries
- Keep only correct entries with proper file paths
- All registry paths now validated and correct
* feat: add add-context command to registry and all profiles
- Register add-context command in registry.json
- Add command:add-context to all 5 profiles (essential, developer, business, full, advanced)
- Update add-context.md to mention external-context files in .tmp/ directory
- Fix command dependencies to use context IDs instead of full paths
- All registry paths and dependencies validated
* feat: enhance add-context with external context file management
- Add Stage 0: Detect external context files in .tmp/
- Offer options to extract, manage, or harvest external files
- Extract patterns from external context files for use in project intelligence
- Provide management options: update, remove, clean, or harvest
- Add examples showing external context extraction and management
- Update troubleshooting with external context file handling
- Integrate with /context harvest command for permanent context organization
* fix: comprehensive context system audit and restructuring
- Fix 6 command documentation inconsistencies in add-context.md
* Correct /add-context vs /context harvest command references
* Simplify Stage 0 workflow from 4 to 2 options
* Update examples to match actual behavior
* Remove non-existent operations from delegation section
* Fix troubleshooting Q&A section
- Restructure registry.json with 4 major improvements
* Replace 31 duplicate 'navigation' IDs with unique IDs per subfolder
- core-navigation, development-navigation, openagents-repo-navigation, etc.
* Add context:openagents-repo/* to developer and full profiles
- Developers now have repo-specific guides and examples
* Remove context:to-be-consumed/* from essential, developer, business, full
- Keep only in advanced profile for system builders
* Properly scope mastra-ai context via wildcard patterns
Validation Results:
- Registry: 220/220 paths valid, 0 missing dependencies
- Profiles: All appropriately scoped with no duplicate IDs
- Standards: MVI compliance, frontmatter, structure verified
- No regressions: All tests passing
Fixes issues with context organization, command consistency, and profile coverage.
* feat: make agents OpenCode-compliant with centralized metadata system
- Remove invalid OpenCode fields from agent frontmatter (id, category, type, version, author, tags, dependencies, context, permissions)
- Keep only valid OpenCode fields (name, description, mode, temperature, tools, permission)
- Create centralized metadata file (.opencode/config/agent-metadata.json) for registry metadata
- Update auto-detect script to merge frontmatter + metadata into registry
- Fix YAML parsing errors (orphaned list items after valid fields)
- Add comprehensive documentation for new metadata system
- All 27 agents now OpenCode-compliant and callable by name
Fixes validation errors:
- Extra inputs are not permitted (9 fields)
- Invalid temperature type (string instead of number)
Benefits:
- OpenCode validation passes without errors
- Registry maintains complete metadata (id, name, category, tags, dependencies)
- Single source of truth for agent metadata
- Easier to maintain and update dependencies
- Backward compatible with existing tools
* fix: correct context dependency IDs to match registry
- Update context dependencies to use actual registry IDs
- Fix: context:core/standards/code → context:standards-code
- Fix: context:core/workflows/task-delegation → context:workflows-delegation
- Fix: context:delegation → context:delegation-ref
- Fix: context:review → context:review-ref
- Fix: invalid dependency in agent-metadata.md documentation
- All registry validation now passes
* chore: add navigation files and update workflow context
- Add navigation.md files for all context directories
- Update design-iteration.md and task-delegation.md workflows
- Clean up backup files from metadata fixes
* fix: install dependencies in validate-registry workflow
The validation script requires the 'glob' package which is in devDependencies.
CI was failing because dependencies were never installed before running validation.
* fix: show validation errors in CI logs
Use tee to display validation output in logs while also saving to file.
This makes debugging CI failures much easier.
* fix: remove duplicate pull_request trigger
The workflow was running twice (pull_request + pull_request_target) causing
duplicate checks. Keep only pull_request_target to support bot PRs and allow
writing to PR branches.
* chore: trigger CI re-run with fixed workflow
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
|
||
|
Darren Hinde
|
c8f7103cb6
|
refactor(evals): consolidate documentation and enhance test infrastructure (#56)
* feat(agents): implement category-based agent organization system - Organize agents into domain categories (core, development, content, data, product, learning) - Move 10 agents to category subdirectories with proper git rename tracking - Update 13 subagents with category and type metadata in frontmatter - Add category metadata files (0-category.json) documenting common patterns - Implement local registry fallback in install script for offline development - Add comprehensive validation suite with 15 automated tests (100% pass rate) - Enhance registry validation with duplicate ID and consistency checks - Update eval framework with intelligent path resolution (backward compatible) - Archive legacy eval structure to _archive/ for reference - Update all documentation to reflect category-based structure - Bump version to 0.5.0 with accurate CHANGELOG Technical Details: - 23 agents organized (10 category agents, 13 subagents) - 6 category directories created - Path resolution supports both agent IDs and category paths - Registry schema updated to v2.0.0 - 159 files changed, 872 insertions(+), 10738 deletions(-) BREAKING CHANGE: Agent file paths now use category structure. Update references from .opencode/agent/openagent.md to .opencode/agent/core/openagent.md. Eval framework maintains backward compatibility via path resolution. * refactor(evals): consolidate documentation and enhance test infrastructure - Remove temporary project tracking files (PHASE_5_COMPLETE, PROJECT_COMPLETE, etc.) - Consolidate evaluation framework docs into main README - Enhance test execution with improved logging and multi-prompt support - Move system-builder from core to meta category - Add comprehensive test suites for openagent with organized structure - Create evaluation test structure for all subagents - Clean up archived workflows and redundant documentation - Update registry to reflect new agent organization - Add shared test templates and golden test patterns * feat(ci): add manual trigger support for bot-created PRs in validate-registry workflow - Add pr_number input to workflow_dispatch for manually triggering validation - Fetch PR details dynamically when triggered manually - Support both automatic (PR event) and manual (workflow_dispatch) triggers - Enable validation of bot-created PRs like automated version bumps - Update branch detection and push logic to handle both trigger types - Add documentation explaining how to manually trigger for bot PRs * feat(evals): add explicit context file validation to test framework - Add expectedContextFiles field to test YAML schema for explicit context file specification - Enhance context-loading-evaluator to support both auto-detect and explicit validation modes - Update documentation with comprehensive guide and examples - Clean up archived legacy test structure (90+ old test files) - Add new example tests demonstrating explicit context validation - Backward compatible with existing tests * feat(evals): add multi-agent logging system and performance optimizations Implement comprehensive multi-agent logging and performance improvements for the eval framework. Task 01: Multi-Agent Logging System - Add complete hierarchical logging module (evals/framework/src/logging/) - SessionTracker: tracks parent-child delegation hierarchies - MultiAgentLogger: pretty-prints logs with visual indentation - Formatters: box characters and emoji formatting - 37 passing unit tests (session-tracker, logger, integration) - Integrate with SDK event stream handler - Hook into session.created, message.updated, message.part.updated events - Real-time child session detection via timestamp heuristics - Message deduplication for cleaner output - Enable in debug mode only (<1% performance overhead) - Add demo script and comprehensive documentation Task 02: Performance Optimizations - Reduce grace period from 5s to 2s (67% reduction, 10-20% faster tests) - Add PerformanceMetricsEvaluator for bottleneck identification - Collects tool latencies, inference time, idle time - Provides full performance visibility - Update smoke test to validate delegation (core feature) - Replace simple read test with multi-agent delegation test - Tests both parent and child agent functionality - Validates multi-agent logging system Results: - 37 unit tests passing - Smoke test passing (95/100 score) - Clean hierarchical logging output - 10-20% faster test execution - Full multi-agent visibility Files changed: 16 files, +2678 lines Time saved: ~6-8 days (completed in 6 hours vs 3-5 day estimate) * feat(evals): show child agent execution in non-debug mode - Enable MultiAgentLogger in both debug and non-debug modes - Add verbose parameter to control output level - Non-verbose mode shows concise child session lifecycle: - Child agent started message - Child agent completed with duration - Verbose mode (--debug) shows full delegation hierarchy - Update documentation to reflect new behavior This provides visibility into delegation without overwhelming output, giving confidence that child agents are actually running. * feat(evals): improve delegation testing and behavior validation Major improvements to eval framework: Test Suite: - Add 00-smoke-test.yaml (basic read operation) - Rename 01-smoke-test.yaml → 02-delegation-test.yaml (delegation test) - Add simple-responder test agent for delegation testing - Add debug test: simple-subagent-call.yaml Evaluators: - Add agent-model-evaluator.ts for agent/model validation - Enhance behavior-evaluator.ts with detailed task tool output - Improve delegation-evaluator.ts with better evidence tracking Schema & Execution: - Add expectedAgent and expectedModel to test schema - Improve test executor with better logging and error handling Documentation: - Update evals/README.md with new features and performance improvements - Remove outdated MULTI_AGENT_LOGGING_COMPLETE.md Registry: - Add simple-responder test agent to registry Fixes: - Fix dashboard.sh path resolution These changes provide better visibility into delegation, improved test coverage, and clearer validation of agent behavior. * chore: add GitHub Actions workflow and plugin documentation - Add .github/workflows/evals/run-evaluations.yml for automated eval testing - Add dev/ai-tools/opencode/plugins/Plugin-inspiration.md for plugin development reference |
||
|
darrenhinde
|
5f2cec1404 |
✨ feat: introduce OpenAgent universal agent with session management
Major architectural refactoring: - Add OpenAgent as universal primary agent for questions, tasks, and workflows - Implement session-based context management with manifest tracking - Reorganize subagents into logical categories (code/core/utils) - Add dynamic context discovery and loading system - Introduce cleanup script for stale sessions - Update registry with new categorization (essential/standard/extended/specialized/meta) - Rename 'core' profile to 'essential' and add 'business' profile - Add comprehensive documentation with visual diagrams - Update installation script with new profiles Breaking changes: - Subagent paths changed from analysis/core to code/core/utils - Profile 'core' renamed to 'essential' (backward compatible) New features: - Session isolation for concurrent agent instances (.tmp/sessions/) - Automatic stale session cleanup (>24 hours) - Manifest-based context indexing and discovery - Business profile for content-focused workflows - OpenAgent with plan-approve-execute-validate-summarize workflow Components: 32 total (3 agents, 13 subagents, 8 commands, 2 tools, 2 plugins, 2 contexts, 2 config) Documentation: - docs/agents/openagent.md - Comprehensive guide with Mermaid diagrams - docs/guides/building-with-opencode.md - Power user formula - scripts/README.md - Script documentation - .opencode/context/core/context-management.md - Session management strategy |
Renamed from .opencode/agent/image-specialist.md (Browse further)