agent-trials
github.com/princejain756/agent-trials- Category
- Developer Tools
- Rank
- No. 3378Tools index
- Pricing
- Open Source
- Type
- TOOL
- Latest release
- v1.0.1
- Date
About
agent-trials runs the same task suite through every coding agent CLI installed on your machine, such as Claude Code, Codex and Gemini CLI, and prints a scoreboard of pass rate, time and cost. Each task runs in a temporary working copy with credentials withheld, and tests the agent cannot edit are restored before verification. It lets you compare agents on your own repo instead of public benchmarks.
Tags
coding-agentbenchmarkevalsclitypescriptclaude-code
Tech Stack
Node.jsTypeScript
Media
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.