by edonadei · Claude Skill · ★ 27
Caliper Reliability testing for agent skills. Define what success looks like, run your skill k times, and get a pass@k score you can track and compare. Agent skills are hard to test. A skill that works on your machine, on this prompt, today, might fail tomorrow after a model update or a one-line prompt edit. Caliper makes reliability measurable: define what success looks like, run the skill repeatedly, and get a pass@k score you can track over time.
| Stars | 27 |
| Forks | 3 |
| Language | Python |
| Category | Claude Skill |
| License | MIT |
| Quality Score | 67.449491953858/100 |
| Open Issues | 7 |
| Last Updated | 2026-07-05 |
| Created | 2026-05-13 |
| Platforms | claude-code, cli, codex, python |
| Est. Tokens | ~17k |
Explore other popular claude skill tools:
caliper is Local-first eval harness for Claude Code, Codex, Hermes and Pi skills. It is categorized as a Claude Skill with 27 GitHub stars.
caliper is primarily written in Python. It covers topics such as ai-agents, claude-code, cli.
You can find installation instructions and usage details in the caliper GitHub repository at github.com/edonadei/caliper. The project has 27 stars and 3 forks, indicating an active community.
caliper is released under the MIT license, making it free to use and modify according to the license terms.