# Changelog

All notable changes to the manage-insights-user skill will be documented in this file.

## [1.1.0] - 2026-06-09

### Changed
- **Renamed skill** from `create-insights-user` to `manage-insights-user` to reflect the broader scope

### Added
- **Copy permissions variant**: New step-by-step guide for copying LPs from one existing user to another (diff-based, no email, no Harness step)
- **Neo4j relationship reference**: Documents `HAS_ACCESS_TO` (external/label users) and `HAS_ADMIN_ACCESS_TO` (employees/admins) relationships, plus Identity lookup pattern
- **Post-run verification step**: After step function completes, query Neo4j to confirm LP counts increased as expected
- **Harness MCP re-test notes**: Updated investigation findings to reflect 2026-06-09 re-test confirming read operations still broken and individual user addition via rules also ruled out

## [1.0.1] - 2026-04-17

### Fixed
- **Step function monitoring**: Updated monitoring logic to use `grep -o -E` to extract only status words, properly handling terminal escape sequences from the wrapper script
- Monitoring now correctly detects SUCCEEDED/FAILED states and exits the polling loop

### Known Issues
- Wrapper script still outputs some terminal escape sequences, but monitoring now filters them correctly
- Future versions will eliminate escape sequences entirely at the wrapper level

## [1.0.0] - 2026-04-16

### Added
- Initial release of the Insights user creation skill
- Full automation from Neo4j query to Harness CSV generation
- Bulk user support - create multiple users in a single invocation
- Automated step function monitoring for both user creation and email workflows
- AWS wrapper script (`aws-with-awsume`) for automatic credential sourcing
- Neo4j integration via `neo4j-prod` MCP server
- CSV generation for:
  - No-email version (user permissions)
  - Send-email version (email invites)
  - Harness segment version (Identity IDs)

### Fixed
- Terminal escape sequences in AWS CLI output by updating wrapper script
- Monitoring reliability by improving status detection logic
- Output filtering to remove noise from awsume and nvm

### Known Limitations
- Harness upload is manual - programmatic API integration planned for future release
- Production only - QA user creation not supported

## Planned for v1.1.0

- **Harness API Integration**: Automate the final step of uploading Identity IDs to Harness segments
- **Better Error Handling**: More detailed error messages and recovery suggestions
- **Dry Run Mode**: Preview changes before executing