Run Codex, Claude Code, Hermes and 28 other agent harnesses behind one OpenAI Responses-compatible API, with sessions, files, streaming and shared memory; self-host or cloud.
Use it now
Keep it with up to 9 other tools you're setting up, then share them all as one link.
Try this first
Ask your coding agent:
- Run Harness Tasks
Use HarnessRouter to run a task on a chosen harness: ‹what you need›
- Idempotent Retries
Use HarnessRouter to get the first result: ‹what you need›
- Task Files
Use HarnessRouter to attach files to a task: ‹what you need›
Call the API
https://api.harnessrouter.ai/v1
Auth A bearer token from HarnessRouter
Setup guide: harnessrouter.ai/docs
Compare
Press Compare to add it to the tray at the bottom of the window, then pick up to 2 others anywhere on the site.
- Access
- API
- Auth
- Bearer Token
- Skills
- 4
Skills
Run Harness Tasks
Runs a task on a chosen harness through the Responses-compatible API, selecting it with metadata.harness_id.
Idempotent Retries
Returns the first result when a request is retried with the same Idempotency-Key, without running the task twice.
Task Files
Attaches files to a task so they land in the agent's working directory before every turn that carries them.
Harness Memory
Keeps long-term memory for each agent that loads at session start and can be shared by agents on different harnesses.