DeepSeek Harness agent framework
fgm-builds/dashr · skills
dsh RLM mode (Recursive Language Models), iPython unified tool-calling interface, context as variables, in dsh Everything is a Plugin ecosystem.
★ 0GitHub stars
0Forks
2026-08-17Last updated
TypeScriptLanguage
MITLicense
Project overview
Instead of paying massive token costs on every round-trip tool call in standard multi-turn chat, Dashr equips the agent with a stateful, persistent Python kernel（持久化内核）. The agent writes self-contained Python programs per cell, manipulating context, tools, and memory as native variables via the Python Kernel Unified Tool Calling（统一的代码化工具调用）. 💡 RLM（Recursive Language Models：递归自调用）
Reference: Recursive Language Models (MIT/Stanford/Open MIND, 2025, arXiv:2512.24601)
The design of Dashr builds upon groundbreaking research in recursive agent execution and persistent prompt harnesses:
Install command
dsh plugin --profile web add github:fgm-builds/dashr