Delegate routes tasks across coding agents

// 1h agoOPENSOURCE RELEASE

Delegate routes tasks across coding agents

Delegate is an open-source plugin for Claude Code that detects installed agent CLIs on a developer's machine and routes tasks to the most suitable harness and model. Supporting 22 harnesses—including Claude inline and sub-agents, Codex, Cursor Agent, OpenCode, Devin, Gemini CLI, Cline, Copilot, OpenHands, and Aider—it evaluates task requirements against model rubrics and user allowlists.

// ANALYSIS

As terminal-based AI tools proliferate into specialized, fragmented CLI harnesses, intelligent routing layers like delegate represent the necessary evolution toward cost-effective, task-specific agent orchestration.

–**Broad harness coverage:** Built-in catalog recognizes 22 harnesses (14 locally verified and 8 documented), distinguishing between file-editing agents, sandboxed runners like Muse, and read-only reviewers like Codex and OpenClaw.
–**Calibrated probabilistic routing:** Employs TypeSafe's Jev model to deliver confidence and probability scores for each allowed candidate, turning potential close calls into transparent tradeoffs.
–**Execution safety:** Decouples selection logic from execution by generating explicit launch strings rather than silently invoking sub-processes with system-level access.
–**Rigorous multi-layer evals:** Ships with dedicated test suites covering plumbing, live API routing benchmarks, and Claude Code plugin trigger evaluations to prevent over-eager activation on routine coding tasks.

// TAGS

delegateclaude-codeagentagent-routingdevtoolopen-sourcellmscli

DISCOVERED

1h ago

2026-09-17

PUBLISHED

2h ago

2026-09-17

RELEVANCE

6/ 10

AUTHOR

mdlahfir

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

MODEL1h ago

Manycore Tech open-sources UniPhysGen for robotic simulation

Manycore Tech, in partnership with Zhejiang University and UESTC, has open-sourced UniPhysGen, a physical AI model that converts raw 3D assets into interactive, simulation-ready digital twins by predicting articulation kinematics and physical properties. Alongside model weights and code, the team released the UniPhys-40K dataset and UniPhys-Bench evaluation benchmark to advance embodied AI research.

RESEARCH1h ago

Schmidhuber traces four decades of recursive self-improvement

AI researcher Jürgen Schmidhuber has published a comprehensive technical note and historical overview titled "Recursive Self-Improvement (RSI) Since 1987," surveying four decades of work in meta-learning and self-modifying computational systems. The retrospective connects foundational paradigms—including self-modifying policies, fast weight programmers, and Gödel Machines—to modern LLM agents, concluding that achieving full artificial superintelligence ultimately requires coupling self-improving software with self-replicating physical hardware in the real world.

INFRA1h ago

ASCII positions boat as cheap Daytona alternative

In an X thread recounting an encounter at Y Combinator, the founder of boat (developed by ASCII) criticized competing cloud sandbox and dev environment providers like Daytona for unreliability and high post-credit pricing, framing boat as a cost-effective, persistent VM alternative for autonomous agents. Engineered for programmatic agent execution with tools like Claude Code, boat provides dedicated Ubuntu VMs with nested Docker support, snapshot forking, and per-second billing granting roughly 555 hours of 4 vCPU compute for $20.

Delegate routes tasks across coding agents