compound-agent 1.0.0 → 1.2.0

package/CHANGELOG.md

@@ -9,6 +9,46 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [1.2.0] - 2026-02-15
+
+### Added
+
+- **`ca loop` command**: Generate autonomous infinity loop scripts that process beads epics end-to-end via chained Claude Code sessions
+- **HUMAN_REQUIRED marker**: Loop detects human-blocking issues, logs reason to beads, skips epic without stopping the loop
+- **Review+compound blocking tasks**: Plan phase now creates review and compound beads issues with dependencies, ensuring these phases survive context compaction and surface via `bd ready`
+
+### Fixed
+
+- **Loop script `set -u` crash**: `LOOP_DRY_RUN` now uses safe expansion (`${VAR:-}`) for `set -u` compatibility
+- **Infinite reprocessing**: Loop tracks processed epics to prevent re-selecting the same epic in dry-run or human-required paths
+- **Input validation**: `--max-retries` rejects non-integer values; epic IDs validated against safe pattern to prevent shell injection
+- **Exit codes**: `ca loop` now returns non-zero on errors (overwrite refusal, invalid options)
+
+## [1.1.0] - 2026-02-15
+
+### Added
+
+- **External reviewer integration**: Optional Gemini CLI and Codex CLI as headless cross-model reviewers. Enable with `ca reviewer enable gemini`. Advisory only, non-blocking.
+- **`ca reviewer` command**: `ca reviewer enable|disable|list` to manage external reviewers
+- **`ca doctor` command**: Verifies external dependencies (bd, embedding model, hooks, MCP server) with actionable fix hints
+- **Dynamic reviewer selection**: Review workflow scales agent count with diff size — 4 core reviewers for small diffs, 7 for medium, full 11 for large
+- **Auto-sync on session start**: `ca prime` now syncs the SQLite index before loading lessons, ensuring MCP searches have fresh data after `git pull`
+- **Config system**: `.claude/compound-agent.json` for user-editable settings (not overwritten by `setup --update`)
+
+### Changed
+
+- **Workflow commands**: Moved slash commands into `compound/` subfolder with legacy cleanup migration
+- **Review template**: Tiered reviewer selection replaces fixed 11-reviewer spawn
+
+### Fixed
+
+- **Safe legacy migration**: `setup --update` and `--uninstall` now check for `GENERATED_MARKER` before deleting root-level commands — user-authored files preserved
+- **Config robustness**: Malformed `.claude/compound-agent.json` no longer crashes reviewer commands
+- **Consistent error handling**: `reviewer disable` and `list` now wrapped in try/catch matching `enable`
+- **Template numbering**: Fixed duplicate step numbers in work.md workflow
+
+## [1.0.0] - 2026-02-15
+
 ### Added
 
 - **Unified memory types**: lesson, solution, pattern, preference -- all share one store, schema, and search mechanism
@@ -398,7 +438,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   - Vitest test suite
   - tsup build configuration
 
-[Unreleased]: https://github.com/Nathandela/learning_agent/compare/v0.2.9...HEAD
+[Unreleased]: https://github.com/Nathandela/learning_agent/compare/v1.2.0...HEAD
+[1.2.0]: https://github.com/Nathandela/learning_agent/compare/v1.1.0...v1.2.0
+[1.1.0]: https://github.com/Nathandela/learning_agent/compare/v1.0.0...v1.1.0
+[1.0.0]: https://github.com/Nathandela/learning_agent/compare/v0.2.9...v1.0.0
 [0.2.9]: https://github.com/Nathandela/learning_agent/compare/v0.2.8...v0.2.9
 [0.2.8]: https://github.com/Nathandela/learning_agent/compare/v0.2.7...v0.2.8
 [0.2.7]: https://github.com/Nathandela/learning_agent/compare/v0.2.6...v0.2.7

package/README.md

@@ -170,6 +170,18 @@ The CLI binary is `ca` (alias: `compound-agent`).
 | `ca rules check` | Run repository-defined rule checks |
 | `ca test-summary` | Run tests and output a compact summary |
 
+### Automation
+
+| Command | Description |
+|---------|-------------|
+| `ca loop` | Generate infinity loop script for autonomous epic processing |
+| `ca loop --epics <ids...>` | Target specific epic IDs |
+| `ca loop -o <path>` | Custom output path (default: `./infinity-loop.sh`) |
+| `ca loop --max-retries <n>` | Max retries per epic on failure (default: 1) |
+| `ca loop --force` | Overwrite existing script |
+
+Generated scripts detect three markers: `EPIC_COMPLETE` (success), `EPIC_FAILED` (retry then stop), `HUMAN_REQUIRED: <reason>` (skip and continue). Run with `LOOP_DRY_RUN=1` to preview.
+
 ### Setup
 
 | Command | Description |