This release adds 3 notable features for engineering teams evaluating rollout.
Published 3mo
AI Agents & Assistants
✓ No known CVEs patched
✓ No known CVEs patched in this version
Topics
agent-loop
agentic-ai-development
agentic-coding
agentic-engineering
agentic-framework
agentic-workflow
+7 more
autonomous-agents
governed-autonomy
governed-loop
harness-engineering
loop-engineering
runtime-engine
runtime-engineering
Summary
AI summaryAdded a specialized skills pipeline mode with configurable Pi event‑log retention.
Full changelog
[0.14.1] - 2026-04-25
0.14.1 packages the repository-local skills pipeline mode alongside a smaller,
more sustainable Pi event-log contract and cleans up asset-policy lint so
millrace doctor reports a clean workspace by default.
Added
- Added the specialized
skills_pipeline_codexbuilt-in mode. - Added
execution.skills_pipelineandplanning.skills_pipelineloop assets. - Added pipeline-specific planning and execution entrypoints for the skills
production flow. - Added integration and asset coverage proving the compiler materializes the
skills pipeline mode contract.
Changed
- Added
runners.pi.event_log_policyso Pi raw event-log retention is
configurable. - Changed Pi runner persistence to keep full raw
runner_eventsonly on failed
runs by default. - Filtered redundant Pi
message_updatesnapshots out of persisted
runner_eventseven when full logging is enabled. - Updated runner and compiler documentation to describe the new skills pipeline
mode and the slimmer Pi event-log behavior.
Fixed
- Fixed asset-policy lint warnings in the
mechanicplanning entrypoint and the
skill-creator reference assets somillrace doctorno longer reports those
false-positive or incomplete-manifest warnings.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About Millrace
All releases →Related context
Related tools
Beta — feedback welcome: [email protected]