# NousResearch/hermes-agent · Autonomous self-improvement and RL fine-tuning proposals

## Open now

- 2 issues, 2 PRs open on GitHub.
- #18092: \[Feature\]: Production-Grade Autonomous Evolution Engine (GASP Loop) for Hermes-Agent
- #61644: \[Feature\]: Autonomous Evaluation and Self-Improvement Engine (HAEE)
- PR #18096: feat: add TrainingDataBridge, export agent trajectories for training
- PR #61645: feat(evolution): add Autonomous Evaluation and Self-Improvement Engine (HAEE)

A cluster of feature requests and proposals around autonomous agent self-improvement, including evaluation engines, reinforcement learning personalization from user feedback, and exporting agent trajectories for training data. These are proposals or additions, not confirmed implementations.

## TL;DR

- 5 member(s): 2 open issue(s), 1 closed issue(s), 2 open PR(s), 0 merged PR(s), 0 PR(s) closed without merging.
- No fix has been merged to main; one or more PRs were attempted (see below).
- **Recommended:** review the prior/competing PR(s) below. A fix was attempted but not merged; your case may extend or revive that work.

## Maintainer resolution verdict

- Internal-tracker verdict over this cluster's issue members: `unresolved`. This is distinct from the fix status above. Weigh it yourself.

## How this cluster was grouped

- Link evidence (`STRUCTURAL_PARTIAL`): some members are connected by structural evidence (shared changed files, or a closing/superseded reference); the rest are joined by an automated relevance judgement over their text.
- Members outside that structural core may belong to a different underlying problem.
- Pairwise link density `0.27`: the share of member pairs that carry a link, counting a structural link in full and a relevance-only link at 0.36. `1.00` means every pair is linked.

## Open related issues

- #18092: \[Feature\]: Production-Grade Autonomous Evolution Engine (GASP Loop) for Hermes-Agent
- #61644: \[Feature\]: Autonomous Evaluation and Self-Improvement Engine (HAEE)

## Closed / duplicate issues

- #498: Feature: Conversational RL Personalization — Online Model Fine-Tuning from Live User Feedback (inspired by OpenClaw-RL)

## Prior & competing PRs

- **#18096: feat: add TrainingDataBridge, export agent trajectories for training**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`; refs `#15426`, `#18092`
  - files: `agent/training_data.py`, `hermes_cli/commands.py`, `hermes_cli/main.py`, `hermes_cli/training_data_cmd.py`, `optional-skills/mlops/haee-train/SKILL.md`, `scripts/benchmark_training_data.py`, `tests/test_training_data.py`
- **#61645: feat(evolution): add Autonomous Evaluation and Self-Improvement Engine (HAEE)**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`
  - files: `AGENTS.md`, `agent/agent_init.py`, `agent/conversation_loop.py`, `agent/evolution/__init__.py`, `agent/evolution/atropos_export.py`, `agent/evolution/auto_export.py`, `agent/evolution/auto_trigger.py`, `agent/evolution/auxiliary_llm.py`, `agent/evolution/config.py`, `agent/evolution/conversation_observer.py`, `agent/evolution/evaluator.py`, `agent/evolution/evolution_hooks.py`, `agent/evolution/evolution_manager.py`, `agent/evolution/evolution_store.py`, `agent/evolution/evolution_tools.py`, `agent/evolution/failure_analyzer.py`, `agent/evolution/harness_variants.py`, `agent/evolution/improvement_metrics.py`, `agent/evolution/improvement_proposer.py`, `agent/evolution/pr_proposer.py` (+29 more)

## Fix status

- No merged fix on record.
- Subsystem: `agent`