compound-agent 1.6.2 → 1.6.4

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md CHANGED
@@ -9,6 +9,28 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
9
9
 
10
10
  ## [Unreleased]
11
11
 
12
+ ## [1.6.4] - 2026-03-07
13
+
14
+ ### Fixed
15
+
16
+ - **Embedding memory leak**: Every `npx ca search` spawned a process that loaded the ~150MB native embedding model. Without explicit cleanup, native threads kept processes alive as zombies. During heavy usage (Claude Code subagents), 32+ processes accumulated = 4.5GB leaked.
17
+ - `withEmbedding(fn)` scoped wrapper guarantees cleanup via try/finally. All 9 command-layer consumers migrated from manual try/finally to this single API.
18
+ - ESLint rule `require-embedding-cleanup` catches any file importing `embedText`/`getEmbedding` without a cleanup function. Scoped to `src/commands/` and `src/setup/`.
19
+ - `cli-app.ts` backstop: top-level finally in `runProgram()` catches anything the above two miss.
20
+ - **`embedText` probe inside `withEmbedding` scope**: `clean-lessons` command was calling `embedText` outside of `withEmbedding`, leaking the model on every invocation.
21
+ - **Agentic skill report format**: Markdown table was missing `|---|` separator row, rendering as plain text in some renderers.
22
+ - **Agentic skill missing setup remediation**: 5 of 15 principles (P8, P12-P15) had no setup actions. Added concrete remediation guidance for each.
23
+ - **Agentic skill missing completion gate**: Other skills have phase gates; the agentic skill was missing one. Added Setup Completion Gate with verification steps.
24
+ - **Agentic skill stack-biased scoring**: Rubric was TypeScript-heavy. Added language-neutral scoring guidance (mypy, clippy, ruff equivalents).
25
+ - **Agentic skill `$ARGUMENTS` dead code**: Mode is set by the calling command (`/compound:agentic-audit` or `/compound:agentic-setup`), not parsed from `$ARGUMENTS`.
26
+ - **Docs template missing agentic commands**: `SKILLS.md` template now lists `agentic-audit` and `agentic-setup` in the command inventory.
27
+
28
+ ## [1.6.3] - 2026-03-05
29
+
30
+ ### Changed
31
+
32
+ - **Cook-it session banner**: Replaced "Claude the Cooker" chef ASCII art with rscr's detailed front-view brain ASCII art ("Claw'd"). Brain features ANSI-colored augmentation zones: cyan neural interface (`##`), bright cyan signal crosslinks (`::`), green bio-circuits (`%`), magenta data highways (`######`), and yellow power nodes (`@@`).
33
+
12
34
  ## [1.6.2] - 2026-03-05
13
35
 
14
36
  ### Fixed
@@ -874,7 +896,12 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
874
896
  - Vitest test suite
875
897
  - tsup build configuration
876
898
 
877
- [Unreleased]: https://github.com/Nathandela/compound-agent/compare/v1.5.0...HEAD
899
+ [Unreleased]: https://github.com/Nathandela/compound-agent/compare/v1.6.4...HEAD
900
+ [1.6.4]: https://github.com/Nathandela/compound-agent/compare/v1.6.3...v1.6.4
901
+ [1.6.3]: https://github.com/Nathandela/compound-agent/compare/v1.6.2...v1.6.3
902
+ [1.6.2]: https://github.com/Nathandela/compound-agent/compare/v1.6.1...v1.6.2
903
+ [1.6.1]: https://github.com/Nathandela/compound-agent/compare/v1.6.0...v1.6.1
904
+ [1.6.0]: https://github.com/Nathandela/compound-agent/compare/v1.5.0...v1.6.0
878
905
  [1.5.0]: https://github.com/Nathandela/compound-agent/compare/v1.4.4...v1.5.0
879
906
  [1.4.4]: https://github.com/Nathandela/compound-agent/compare/v1.4.3...v1.4.4
880
907
  [1.4.3]: https://github.com/Nathandela/compound-agent/compare/v1.4.2...v1.4.3
package/README.md CHANGED
@@ -122,6 +122,16 @@ Or run all phases sequentially:
122
122
  /compound:cook-it "Add auth to API"
123
123
  ```
124
124
 
125
+ Additional commands:
126
+
127
+ ```
128
+ /compound:learn-that --> Capture a lesson from conversation context
129
+ /compound:check-that --> Search lessons and apply to current work
130
+ /compound:get-a-phd --> Deep research to build agent knowledge
131
+ /compound:agentic-audit --> Score codebase against agentic manifesto
132
+ /compound:agentic-setup --> Audit then set up agentic infrastructure
133
+ ```
134
+
125
135
  Each phase searches memory for relevant past knowledge and injects it into agent context. The compound phase captures new knowledge, closing the loop.
126
136
 
127
137
  ## CLI Reference