HarnessRouter

Navigate HarnessRouter

Search the site

  • HarnessRouter Community EditionOpen-source agent harness infrastructure and the Unified Harness Protocol.Product
  • Harness catalogSupported harness IDs and source-backed integration references.Harnesses
  • Care Prep BenchmarkThe same task compared across Codex, Claude Code, and Hermes configurations.Benchmarks
  • Harness rankingsPublic attributed usage across reviewed agent harnesses.Product
  • ExamplesSee how agent-backed features return finished work inside a product.Product
  • PricingPlans and usage pricing for HarnessRouter.Product
  • Hermes AgentAn open-source, persistent AI agent with memory, reusable skills, built-in tools, browser automation, scheduled tasks, and subagents.Harnesses
↑↓ Navigate↵ OpenEsc Close
Sign up

Claude Code

By Anthropic · Available on HarnessRouter

Run itCompare in rankings

Anthropic’s coding agent for understanding codebases, editing files, running commands and tests, and iterating on implementation tasks.

30-day usageAug 14–Sep 12, 2026

Data as of September 13, 2026 · About the data

Rank#2#2 of 21 CLI agents
Tokens17.8T
Requests172.9M

Capabilities and verification

Reviewed against the HarnessRouter capability catalog on August 24, 2026. Every verified row cites its public source; rule-derived rows apply HarnessRouter's published classification rules to documented behavior. A “No public evidence” badge marks a gap in the public record, not a statement that Claude Code lacks them.

Configuration & support

  • Other modelsNo public evidence
  • Custom instructionsDirectly verified
  • SkillsDirectly verified
  • Custom toolsDirectly verified
  • MCPDirectly verified
  • Sandbox & approvalsDirectly verified
  • SessionsDirectly verified
  • SubagentsDirectly verified

Execution & agency

  • Workspace filesDirectly verified
  • Structured actionsDirectly verified

Environment control

  • Browser controlDirectly verified
  • Computer controlDirectly verified

Observation & capture

  • Screen observationDirectly verified
  • Screenshot exportNo public evidence
  • Recording exportDirectly verified

Artifact outputs

  • DocumentsRule-derived
  • SpreadsheetsRule-derived
  • PresentationsRule-derived
  • ImagesRule-derived
  • VideoRule-derived
  • App deploymentNo public evidence

Sources: code.claude.com

Machine interfaces

  • Headless CLI · Stable
  • SDK · Stable

Benchmark evidence

Claude Code configurations from the HarnessRouter Care Prep benchmark, published August 10, 2026. One controlled synthetic-data task; results apply to this test, not universally.

ModelEnd-to-end latencyCost
claude-sonnet-4.61m 31s83.5 credits
claude-opus-4.83m 18s223 credits

Execution success and strict grounding pass rates for these configurations are published in the observed-history capture on the benchmark report.

See the full benchmark report

Run Claude Code through HarnessRouter

API harness ID: claude-code

Use it for code changes or finished documents that your product can review.

Create an API key and run your first task

Run a taskHarnesses API

Nearest CLI agents by usage rank in the same 30-day window.

Run Claude Code through one API.

Ship it behind your product and compare it on your own tasks. Start free with your own model keys, no card needed.

Start building free