claudomator.git - claudomator — task automation server

Age	Commit message (Collapse)	Author
5 days	fix(executor): validate the model classifier's output against a known model list	Peter Stone
	The Gemini-based Classifier trusted its LLM output verbatim. Twice in production it echoed part of its own prompt's JSON schema literally -- 'model-name' (the prompt's own placeholder, also fixed here to be less echo-prone) and separately 'choose-the-best-model' -- instead of substituting a real model identifier. Both were syntactically valid JSON strings that passed straight through to the claude CLI's --model flag, which rejected them and failed the task outright. validateClassification now rejects anything outside the known model list, which Classify's existing error handling already treats as 'classification failed' and falls back to no explicit model override -- the same documented fallback path, now also covering 'succeeded but returned garbage.'
2026-04-28	feat(executor): add LocalRunner and OpenAI-compat LLM client	Claude
	Phase 1 of "local OSS models as agents" plan. Adds a third Runner backed by any OpenAI-compatible HTTP server (Ollama, vLLM, LM Studio, llama.cpp), and migrates the Gemini-CLI classifier to route through the same client when configured. Two-layer split: internal/llm.Client is the workhorse (HTTP, no Pool, no DB) used directly by the classifier and any future internal helper that needs cheap reasoning. internal/executor.LocalRunner is a thin adapter implementing Runner for user-facing tasks. This avoids Pool reentrancy/deadlock when sub-second internal calls fire from inside Pool.execute(). Highlights: - internal/retry: relocated runWithBackoff/IsRateLimitError/ParseRetryAfter into a shared package reused by executor and llm. - internal/llm: Chat (non-streaming) and ChatStream (SSE) over /chat/completions with optional bearer auth, json_object response format, retry on 429/503, Retry-After parsing. - internal/executor/LocalRunner: streams deltas into stdout.log in the same stream-json envelope ClaudeRunner emits, then writes one consolidated assistant block plus a result terminator so existing parsers (extractSummary, ParseChangestatFromOutput) work unchanged. - internal/executor/Classifier: gains optional LLM field; uses json_object response format (no markdown-fence cleanup needed). Falls back to Gemini-CLI subprocess when LLM is nil. - Pool.skipClassification: now skips only when the requested agent type is registered, so unknown types still reach the load balancer. - Storage: additive tokens_in/tokens_out ALTERs on executions; CLI runners record cost_usd as before, LocalRunner records 0 + tokens. - Config: [local_model] section (endpoint, model, timeout_seconds, default_temperature, api_key). Empty endpoint = no LocalRunner registered, classifier falls back to Gemini. Pre-existing test issues fixed in passing: - claude_test.go setupSandbox callsites updated to current signature. - gemini_test.go TestParseGeminiStream skipped (asserts unimplemented GeminiRunner stream-error parsing; tracked separately). Plan: docs/plans/local-oss-runner.md. https://claude.ai/code/session_017Edeq947TpSm1vQTxMhi1J
2026-03-10	executor: explicit load balancing — code picks agent, classifier picks model	Peter Stone
	pickAgent() deterministically selects the agent with the fewest active tasks, skipping rate-limited agents. The classifier now only selects the model for the pre-assigned agent, so Gemini gets tasks from the start rather than only as a fallback when Claude's quota is exhausted. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-09	executor: strengthen rate-limit avoidance in classifier	Peter Stone
	Updated isQuotaExhausted to detect more Claude quota messages. Added 'rate limit reached (rejected)' to quota exhausted checks. Strengthened classifier prompt to explicitly forbid selecting rate-limited agents. Improved Pool to set 5h rate limit on quota exhaustion.
2026-03-09	executor: update classifier model list to Claude 4.x and current Gemini models	Peter Stone

2026-03-09	executor: fix Claude rate-limit detection and prioritize Gemini when limited	Peter Stone
	Updated parseStream to detect 'rate_limit_event' and 'assistant' error:rate_limit messages from the Claude CLI. Updated Classifier to strongly prefer non-rate-limited agents. Added logging to Pool to track rate-limit status during classification.
2026-03-08	executor: update gemini model to 2.5-flash-lite and fix classifier parsing	Peter Stone
	Update the default Gemini model and classification prompt to use gemini-2.5-flash-lite, which is the current available model. Improved the classifier's parsing logic to correctly handle the JSON envelope returned by the gemini CLI (stripping 'response' wrapper and 'Loaded cached credentials' noise).
2026-03-08	feat(executor): implement Gemini-based task classification and load balancing	Peter Stone
	- Add Classifier using gemini-2.0-flash-lite to automatically select agent/model. - Update Pool to track per-agent active tasks and rate limit status. - Enable classification for all tasks (top-level and subtasks). - Refine SystemStatus to be dynamic across all supported agents. - Add unit tests for the classifier and updated pool logic. - Minor UI improvements for project selection and 'Start Next' action.