cat-shark/pi-commandcode-provider
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(models): move manual reasoning efforts into a catalog override

Browse Source 
PR #69 added efforts for meta/muse-spark-* directly to the generated
catalog, which the drift check flags and the daily sync job reverts.

Keep src/commandcode-catalog.ts byte-identical to upstream and merge a
separate src/commandcode-catalog-overrides.ts over it at load time. A
test fails as soon as upstream publishes efforts for an overridden model
so the override gets removed instead of shadowing the CLI catalog.

Verified against the live endpoint: pi --thinking xhigh sends
reasoning_effort="xhigh" for meta/muse-spark-1.2-contributor and the
request succeeds; --thinking off sends none.

Closes #69
This commit is contained in:
Patrick Wozniak
2026-09-02 00:03:45 +02:00
parent df4f4a6eef
commit d3c9832d28
7 changed files with 61 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+1 -1
View File
@@ -88,7 +88,7 @@ Open `/model` and select one of the models provided by Command Code. Model avail
### Reasoning support
Reasoning capability and selectable effort levels follow the official CLI catalog independently. Models can therefore be marked as reasoning-capable even when Command Code chooses their depth automatically; for those models (for example `meta/muse-spark-1.1`, `meta/muse-spark-1.2`, and `meta/muse-spark-1.2-contributor`) the thinking level stays `off` because the CLI itself sends no effort parameter and Command Code has not published selectable levels. Models with explicit effort support register a model-specific `thinkingLevelMap`, so pi and OMP expose only valid levels. Pi's native OpenAI- and Anthropic-compatible providers translate the selected level for Provider API accounts; the existing Command Code generate transport sends the matching `reasoning_effort` for Go accounts.
Reasoning capability and selectable effort levels follow the official CLI catalog independently. Models can therefore be marked as reasoning-capable even when Command Code chooses their depth automatically. Models with explicit effort support register a model-specific `thinkingLevelMap`, so pi and OMP expose only valid levels. For a few reasoning models the CLI catalog ships no effort levels although the endpoint accepts `reasoning_effort`; `src/commandcode-catalog-overrides.ts` adds a manual level set for those (currently `meta/muse-spark-1.1`, `meta/muse-spark-1.2`, and `meta/muse-spark-1.2-contributor`) on top of the generated catalog, and the tests fail once upstream publishes its own levels so the override gets removed. Pi's native OpenAI- and Anthropic-compatible providers translate the selected level for Provider API accounts; the existing Command Code generate transport sends the matching `reasoning_effort` for Go accounts.
List Command Code models from the terminal: