This release fixes issues for SREs watching stability and regressions.
✓ No known CVEs patched in this version
Topics
+6 more
Summary
AI summaryBroken user tools degrade gracefully instead of crashing the agent.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Feature | Medium |
Broken user tools degrade gracefully instead of crashing the agent. Broken user tools degrade gracefully instead of crashing the agent. Source: granite4.1:8b-q6_K@2026-05-19 Confidence: high |
— |
| Feature | Medium |
Calling a broken user‑tool stub returns the error message for diagnosis. Calling a broken user‑tool stub returns the error message for diagnosis. Source: granite4.1:30b@2026-05-19-audit Confidence: low |
— |
| Bugfix | Low |
User tools that fail to import become stubs with `[broken]` in description. User tools that fail to import become stubs with `[broken]` in description. Source: granite4.1:30b@2026-05-19-audit Confidence: low |
— |
Full changelog
Broken user tools degrade gracefully instead of crashing the agent.
When a tool in ~/.mi/tools/ fails to import (missing package, syntax error, etc.), it becomes a stub with [broken] in its description. Calling the stub returns the error message, so the model can diagnose and self-fix — install the missing dep, edit the file, then retry on the next loop iteration.
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 Tiny agentic loop with Docker sandbox
All releases →Related context
Related tools
Earlier breaking changes
- v1.8.0 Goal tool's `check` parameter now a judge prompt, not bash command.
Beta — feedback welcome: [email protected]