This release adds 2 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+13 more
Summary
AI summaryAutomatic model selection based on task complexity across seven engine families.
Full changelog
Autonomous model selection + portable governance
martin run picks the right model automatically
No --model flag needed. MartinLoop classifies task complexity:
- Simple task → haiku tier (claude-haiku, gpt-4o-mini, gemini-2.5-flash)
- Architectural work → sonnet tier
- Security + migration → opus tier (claude-opus, o3, gemini-2.5-ultra)
Works across all 7 engine families: Claude, Codex, Gemini, DeepSeek, Qwen, Llama, Nemotron.
Portable on every machine
All hooks use npx martin-loop gate — no global install required. Paths use homedir() and path.join() — works on macOS, Linux, and Windows.
28 new tests
12 memory store tests + 16 model auto-selection tests. All real, no stubs.
Install
npm install -g [email protected]
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 Keesan12/Martin-Loop
All releases →Related context
Related tools
Earlier breaking changes
- vmcp-v0.1.3 martin_status uses oneOf for selector exclusivity, latest as const.
- vmcp-v0.1.3 maxIterations and maxTokens modeled as integers in tool schemas.
- vmcp-v0.1.3 Tool schemas enforce additionalProperties: false on public contracts.
- vmcp-v0.1.3 Packaged artifacts now require and ship server.json alongside package.json.
Beta — feedback welcome: [email protected]