mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-05-02 02:01:47 +00:00
- AgentResult now includes tool-call metrics: attempted, schema_valid, executed_ok, exec_error - HermesAgentLoop normalizes args robustly without crashing, but distinguishes schema-valid args (dict) from coerced formats (stringified JSON, plain strings) - SweSmithOracleEnv reward shaping now prefers schema-valid tool calls while still giving small credit for attempted tool use |
||
|---|---|---|
| .. | ||
| configs | ||
| tool_call_parsers | ||
| __init__.py | ||
| agent_loop.py | ||
| gsm8k_agent_env.py | ||
| hermes_base_env.py | ||
| hermes_swe_env.py | ||
| patches.py | ||
| swe_smith_oracle_env.py | ||
| terminal_test_env.py | ||
| tool_context.py | ||