open-agents-ai 0.42.1 → 0.44.0

package/README.md

@@ -37,6 +37,7 @@ An autonomous multi-turn tool-calling agent that reads your code, makes changes,
 - **Sub-agent delegation** — spawn independent agents for parallel workstreams
 - **Ralph Loop** — iterative task execution that keeps retrying until completion criteria are met
 - **Dream Mode** — creative idle exploration modeled after real sleep architecture (NREM→REM cycles)
+- **Autoresearch Swarm** — 5-agent GPU experiment loop during REM sleep: Researcher, Monitor, Evaluator, Critic, Flow Maintainer autonomously run ML training experiments, keep improvements, discard regressions
 - **Live Listen** — bidirectional voice communication with real-time Whisper transcription
 - **Neural TTS** — hear what the agent is doing via GLaDOS or Overwatch ONNX voices, with personality-driven expressiveness
 - **Personality Core** — SAC framework-based style control (concise/balanced/verbose/pedagogical) that shapes agent response depth, voice expressiveness, and system prompt behavior
@@ -293,6 +294,47 @@ Each cycle expands through all four stages then contracts (evaluation, pruning o
 
 All proposals are indexed in `.oa/dreams/PROPOSAL-INDEX.md` for easy review.
 
+### Autoresearch Swarm — 5-Agent GPU Experiment Loop
+
+When a GPU is detected and the model tier is "large", the REM stage of Dream Mode activates the **Autoresearch Swarm** instead of the standard multi-agent creative exploration. This is a 5-agent system inspired by [Karpathy's autoresearch](https://github.com/karpathy/autoresearch) that autonomously runs ML training experiments.
+
+The swarm operates in four phases:
+
+| Phase | What Happens |
+|-------|-------------|
+| **Phase 0: Load** | Reads autoresearch memory (best config, experiment log, failed approaches, hypothesis queue, architectural insights) + detects GPU specs |
+| **Phase 1: Hypothesis** | Critic generates 5-8 hypotheses; Flow Maintainer plans experiment ordering and round budget |
+| **Phase 2: Experiment** | Sequential rounds (up to 3): Critic pre-screens → Researcher modifies train.py + runs → Monitor watches GPU → Evaluator keeps/discards → Flow Maintainer decides continue/stop |
+| **Phase 3: Summary** | Flow Maintainer writes consolidated summary to memory + dream report to `.oa/dreams/` |
+
+#### The 5 Agent Roles
+
+| Role | MaxTurns | Temp | Purpose |
+|------|----------|------|---------|
+| **Researcher** | 25 | 0.4 | Modifies train.py, runs experiments via `autoresearch` tool |
+| **Monitor** | 5 | 0.1 | Watches GPU utilization, reports status (detachable between rounds) |
+| **Evaluator** | 12 | 0.3 | Compares results to best val_bpb, calls keep/discard, writes insights to memory |
+| **Critic** | 8 | 0.5 | Generates hypotheses, pre-screens before GPU time is spent |
+| **Flow Maintainer** | 10 | 0.3 | Orchestrates rounds, manages hypothesis queue, writes final summary |
+
+#### Bidirectional Memory
+
+The swarm maintains persistent memory in `.oa/memory/autoresearch.json` with five keys:
+
+- **best_config** — best val_bpb and what train.py changes produced it
+- **experiment_log** — chronological list of experiments with hypotheses, results, and verdicts
+- **architectural_insights** — patterns learned (what architectures work, what doesn't)
+- **failed_approaches** — things NOT to try again (with reasons)
+- **hypothesis_queue** — pending ideas for future experiments
+
+Memory flows bidirectionally: the swarm reads all 5 keys at startup (Phase 0) and writes results back after each experiment. The DMN's gather phase naturally discovers autoresearch learnings when searching all memory, and DMN proposals with category `"autoresearch"` execute through the normal agentic loop.
+
+#### Monitor Detachability
+
+The Monitor agent can be "detached" between experiment rounds by the Flow Maintainer. When detached, the monitor receives a sub-task (e.g., "analyze GPU memory patterns from last 3 runs") instead of its standard watch prompt. This lets the swarm use idle monitoring capacity for useful analysis work.
+
+If no GPU is detected, the REM stage falls back to the standard multi-agent creative exploration (Visionary + Pragmatist + Cross-Pollinator + Synthesizer).
+
 ## Blessed Mode — Infinite Warm Loop
 
 `/full-send-bless` activates an infinite warm loop that keeps model weights loaded in VRAM and the agent ready for instant response. The engine sends periodic keep-alive pings to the inference backend (every 2 minutes) to prevent Ollama's automatic model unloading.
@@ -322,7 +364,7 @@ Inspired by the brain's Default Mode Network (Raichle 2001), the DMN activates d
 
 Each DMN cycle runs a lightweight LLM agent (15 max turns, temperature 0.4) with read-only file access plus full memory tools. The DMN writes insights back to memory, creating a self-reinforcing knowledge loop.
 
-**Task categories**: directive (standing orders), exploration (knowledge gaps), capability (underused tools), maintenance (system health), social (communication)
+**Task categories**: directive (standing orders), exploration (knowledge gaps), capability (underused tools), maintenance (system health), social (communication), autoresearch (autonomous GPU ML experiment loop)
 
 **Backoff**: After 3 consecutive cycles with no actionable task, the DMN enters extended rest. A 30-second cooldown between null cycles prevents spin-looping.