open-agents-ai 0.187.6 → 0.187.7

package/README.md

@@ -2943,7 +2943,7 @@ No configuration needed — the cascade is built from your endpoint usage histor
 
 <div align="right"><a href="#top">back to top</a></div>
 
-46 evaluation tasks test the agent's autonomous capabilities across coding, web research, SDLC analysis, tool creation, multi-file reasoning, memory systems, and context engineering:
+234+ evaluation tasks test the agent's autonomous capabilities across coding, web research, SDLC analysis, tool creation, multi-file reasoning, memory systems, context engineering, multi-agent orchestration, and browser automation:
 
 ```bash
 node eval/run-agentic.mjs                          # Run all tasks
@@ -3098,6 +3098,42 @@ node eval/web-nav/run-web-nav.mjs yadaphone-rates --model qwen3.5:9b
 
 Research papers applied: [AgentOccam](https://arxiv.org/abs/2410.13825) (ICLR 2025), [D2Snap](https://arxiv.org/abs/2508.04412), [Mind2Web](https://arxiv.org/abs/2306.06070) (NeurIPS 2023), [SeeAct](https://arxiv.org/abs/2401.01614), [Fara-7B](https://arxiv.org/abs/2511.19663), [Agent-E](https://arxiv.org/abs/2407.13032), [V-GEMS](https://arxiv.org/abs/2603.02626), [Building Browser Agents](https://arxiv.org/abs/2511.19477), [WebAgent-R1](https://arxiv.org/abs/2505.16421) (EMNLP 2025), [WebRL](https://arxiv.org/abs/2411.02337) (ICLR 2025).
 
+### Multi-Agent Architecture Evaluation (v0.187.4)
+
+43 tasks across 8 categories testing the Hannover-aligned agent spawning system: typed agents (general/explore/plan/coordinator), parallel delegation, inter-agent messaging, worktree isolation, and multi-step orchestration pipelines.
+
+```bash
+node eval/run-agentic.mjs ma-explore-01     # Single agent task
+node eval/run-agentic.mjs ma-triage         # Run a category
+node eval/run-agentic.mjs --model qwen3.5:4b  # Different model tier
+```
+
+**Literature grounding** (11 papers, 2023-2026): [AgentVerse](https://arxiv.org/abs/2308.10848) (4-stage recruit/decide/execute/evaluate), [MASS](https://arxiv.org/abs/2502.11578) (multi-agent topology optimization), [OpenHands](https://arxiv.org/abs/2511.03690) (sandboxed agent SDK), [SWE-bench](https://arxiv.org/abs/2310.06770) (real GitHub issue resolution), [ExpeL](https://arxiv.org/abs/2308.10144) (experiential learning), [Sol-Ver](https://arxiv.org/abs/2502.14948) (solver-verifier self-play), [SPELL](https://arxiv.org/abs/2509.23863) (3-role competitive self-play), [tau-bench](https://arxiv.org/abs/2406.12045) (pass^k reliability), [LatentMAS](https://arxiv.org/abs/2511.20639) (latent collaboration), [Incident Response](https://arxiv.org/abs/2511.15755) (80x specificity from multi-agent), [EvoSkill](https://arxiv.org/abs/2603.02766) (automated skill evolution).
+
+**Results by category (9B model):**
+
+| Category | Pattern | Tasks | Pass Rate |
+|----------|---------|-------|-----------|
+| ma-explore | Explore agent finds issues, general agent fixes | 5 | 4/5 (80%) |
+| ma-triage | 3-5 parallel agents fix independent bugs | 5 | 5/5 (100%) |
+| ma-web | Web scrape data, synthesize into code modules | 5 | 5/5 (100%) |
+| ma-refactor | Multi-file architecture pattern extraction | 5 | 5/5 (100%) |
+| ma-research | Web research, then implement from findings | 5 | 5/5 (100%) |
+| ma-verify | Plan agent designs, general implements, explore verifies | 5 | 5/5 (100%) |
+| ma-compete | Two agents solve independently, best solution selected | 5 | 5/5 (100%) |
+| ma-feature | Long-horizon multi-file feature builds with verification | 5 | 5/5 (100%) |
+| **Total** | | **40** | **39/40 (97.5%)** |
+
+**Cross-model results:**
+
+| Model | Tier | Tasks Run | Pass Rate |
+|-------|------|-----------|-----------|
+| qwen3.5:4b | small | 8 representative | 7/8 (87.5%) |
+| qwen3.5:9b | medium | 40 full suite | 39/40 (97.5%) |
+| qwen3.5:27b | large | 8 representative | 8/8 (100%) |
+
+**Agent architecture components tested:** agent type registry (4 types), per-type tool filtering (allowlist/denylist), unified `agent` tool with `subagent_type` parameter, `send_message` inter-agent communication, `enter_worktree`/`exit_worktree` git isolation, background agent spawning with `run_in_background`, coordinator mode with worker limits.
+
 ### REST API Enterprise Evaluation (v0.185.68)
 
 35 test cases executed against the oa REST API (`oa serve` on port 11435) across **10 industries** and **3 model tiers**. Each case sends a domain-specific prompt via `/v1/chat/completions` and verifies correctness against expected patterns.