* refactor: Reduce continuation prompt token usage by 95 lines
Removed redundant instructions from continuation prompt that were originally
added to mitigate a session continuity issue. That issue has since been
resolved, making these detailed instructions unnecessary on every continuation.
Changes:
- Reduced continuation prompt from ~106 lines to ~11 lines (~95 line reduction)
- Changed "User's Goal:" to "Next Prompt in Session:" (more accurate framing)
- Removed redundant WHAT TO RECORD, WHEN TO SKIP, and OUTPUT FORMAT sections
- Kept concise reminder: "Continue generating observations and progress summaries..."
- Initial prompt still contains all detailed instructions
Impact:
- Significant token savings on every continuation prompt
- Faster context injection with no loss of functionality
- Instructions remain comprehensive in initial prompt
Files modified:
- src/sdk/prompts.ts (buildContinuationPrompt function)
- plugin/scripts/worker-service.cjs (compiled output)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor: Enhance observation and summary prompts for clarity and token efficiency
* Enhance prompt clarity and instructions in prompts.ts
- Added a reminder to think about instructions before starting work.
- Simplified the continuation prompt instruction by removing "for this ongoing session."
* feat: Enhance settings.json with permissions and deny access to sensitive files
refactor: Remove PLAN-full-observation-display.md and PR_SUMMARY.md as they are no longer needed
chore: Delete SECURITY_SUMMARY.md since it is redundant after recent changes
fix: Update worker-service.cjs to streamline observation generation instructions
cleanup: Remove src-analysis.md and src-tree.md for a cleaner codebase
refactor: Modify prompts.ts to clarify instructions for memory processing
* refactor: Remove legacy worker service implementation
* feat: Enhance summary hook to extract last assistant message and improve logging
- Added function to extract the last assistant message from the transcript.
- Updated summary hook to include last assistant message in the summary request.
- Modified SDKSession interface to store last assistant message.
- Adjusted buildSummaryPrompt to utilize last assistant message for generating summaries.
- Updated worker service and session manager to handle last assistant message in summarize requests.
- Introduced silentDebug utility for improved logging and diagnostics throughout the summary process.
* docs: Add comprehensive implementation plan for ROI metrics feature
Added detailed implementation plan covering:
- Token usage capture from Agent SDK
- Database schema changes (migration #8)
- Discovery cost tracking per observation
- Context hook display with ROI metrics
- Testing and rollout strategy
Timeline: ~20 hours over 4 days
Goal: Empirical data for YC application amendment
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* feat: Add transcript processing scripts for analysis and formatting
- Implemented `dump-transcript-readable.ts` to generate a readable markdown dump of transcripts, excluding certain entry types.
- Created `extract-rich-context-examples.ts` to extract and showcase rich context examples from transcripts, highlighting user requests and assistant reasoning.
- Developed `format-transcript-context.ts` to format transcript context into a structured markdown format for improved observation generation.
- Added `test-transcript-parser.ts` for validating data extraction from transcript JSONL files, including statistics and error reporting.
- Introduced `transcript-to-markdown.ts` for a complete representation of transcript data in markdown format, showing all context data.
- Enhanced type definitions in `transcript.ts` to support new features and ensure type safety.
- Built `transcript-parser.ts` to handle parsing of transcript JSONL files, including error handling and data extraction methods.
* Refactor hooks and SDKAgent for improved observation handling
- Updated `new-hook.ts` to clean user prompts by stripping leading slashes for better semantic clarity.
- Enhanced `save-hook.ts` to include additional tools in the SKIP_TOOLS set, preventing unnecessary observations from certain command invocations.
- Modified `prompts.ts` to change the structure of observation prompts, emphasizing the observational role and providing a detailed XML output format for observations.
- Adjusted `SDKAgent.ts` to enforce stricter tool usage restrictions, ensuring the memory agent operates solely as an observer without any tool access.
* feat: Enhance session initialization to accept user prompts and prompt numbers
- Updated `handleSessionInit` in `worker-service.ts` to extract `userPrompt` and `promptNumber` from the request body and pass them to `initializeSession`.
- Modified `initializeSession` in `SessionManager.ts` to handle optional `currentUserPrompt` and `promptNumber` parameters.
- Added logic to update the existing session's `userPrompt` and `lastPromptNumber` if a `currentUserPrompt` is provided.
- Implemented debug logging for session initialization and updates to track user prompts and prompt numbers.
---------
Co-authored-by: Claude <noreply@anthropic.com>
Improvements:
- Enhanced null handling for empty/whitespace fields
- Ensures clean null values in database instead of empty strings
- Improves query efficiency and data consistency
Testing:
- Added comprehensive regression test suite (49 tests)
- Tests v4.2.5 summary fixes and v4.2.6 observation fixes
- Tests edge cases: missing fields, empty fields, whitespace
- New test script: npm run test:parser
- All tests passing with 100% coverage
Code Quality:
- Removed unused extractFileArray() function
- Improved function documentation
- TypeScript diagnostics clean
Technical Details:
- Updated src/sdk/parser.ts extractField function
- Created src/sdk/parser.test.ts regression test suite
- Updated package.json to v4.2.7
- Updated CLAUDE.md with version history
- All changes backward compatible
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Critical Bugfix:
- Fixed overly defensive observation validation blocking observations from being saved
- Parser now NEVER skips observations - always saves them
- Invalid or missing type defaults to "change" (generic catch-all type)
- Removed validation requiring title, subtitle, and narrative fields
- Prevents critical data loss - partial observations better than no observations
Impact:
- Before: Missing title, subtitle, OR narrative caused entire observation to be discarded
- After: ALL observations preserved regardless of field completeness
- Even partial observations contain valuable data: concepts, files_read, files_modified, facts
- LLMs make mistakes - system must be resilient and save everything
- Consistent with v4.2.5 summary fix
Technical changes:
- Updated src/sdk/parser.ts:52-67 to never skip observations
- Uses "change" as fallback type for invalid/missing types (no schema change)
- Updated ParsedObservation interface to allow null for title, subtitle, narrative
- Updated SessionStore.storeObservation signature to accept nullable fields
- Updated built worker-service.cjs
- Bumped version to 4.2.6 in all metadata files
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Removed validation for required fields in parseSummary function.
- Added a note emphasizing the importance of saving the summary even if some fields are missing.
Added "WHEN NOT TO SUMMARIZE" section to buildSummaryPrompt that instructs the SDK to skip creating summaries for:
- Work already covered in previous prompts (prevents duplicates)
- Conversational banter with no deliverables
- Trivial requests (questions, status checks)
- Meta-discussions about memory system without shipped changes
Implementation:
- src/sdk/prompts.ts: Added WHEN NOT TO SUMMARIZE section with <skip_summary> output format
- src/sdk/parser.ts: Added skip_summary detection before parsing full summary XML
- src/sdk/parser.ts: Fixed observation type validation to include all 6 types (bugfix, feature, refactor, change, discovery, decision)
This should eliminate the duplicate summaries like the three "restore 6 types" summaries we saw for session d9137878.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Filter out observation type from concepts array in parseObservations function to ensure types and concepts are treated as separate dimensions. Added logging for removed types.
- Update prompts documentation to clarify that the observation type must not be included in the concepts array.
- Modify search-server to provide clearer guidance on result limits, emphasizing starting with smaller limits to avoid exceeding token limits.
- Refactor SessionSearch methods to accept options for limit, offset, and orderBy parameters, improving flexibility in querying observations by concept and type.
- Introduced a new Logger utility to standardize logging with correlation IDs and structured context.
- Replaced console.error and console.log statements with logger methods in various modules including save.ts, summary.ts, parser.ts, HooksDatabase.ts, and worker-service.ts.
- Enhanced error handling and logging for better traceability of observations and summaries.
- Made observations.text nullable in the database schema to support structured fields.
- Added correlation IDs for tracking observations through the processing pipeline.
- Updated observation schema to include hierarchical fields: title, subtitle, facts, narrative, concepts, files_read, and files_modified.
- Modified the save-hook and summary-hook scripts to accommodate the new observation structure.
- Added migration logic to the HooksDatabase for adding new fields to the observations table.
- Refactored the parser to extract new fields from XML formatted observations.
- Adjusted prompt generation to reflect the new observation format and requirements.
- Updated worker service to handle new observation and summary structures.
- Added background agent architecture for processing tool observations and generating session summaries.
- Created SDK Prompts Module for generating prompts for the Claude Agent SDK.
- Developed XML Parser Module for parsing observation and summary XML blocks from SDK responses.
- Implemented SDK Worker Process to handle observation processing and session management.
- Updated newHook implementation to spawn the SDK worker as a detached process with path resolution for development and production.
- Created comprehensive test suite for SDK prompts, XML parsing, and HooksDatabase integration, ensuring all tests pass.
- Documented Phase 2 implementation details, architecture validation, and success criteria in PHASE2-COMPLETE.md.
Alex Newman