Back to directory
Testing & Reviewqualixar/fidelitybench
fidelitybench
A communication fidelity benchmark for multi-agent AI systems. Measuring per-hop information degradation in LLM agent chains.
Suggested install command
npx skills add qualixar/fidelitybench/fidelitybenchAlways inspect the linked repository and skill instructions before running commands. Skills are instructions; permissions and execution still matter.
Compatibility
Agent support matrix
3 supported
| Agent | Status |
|---|---|
| Claude Code | Supported |
| OpenCode | Not listed |
| Cursor | Supported |
| MCP | Not listed |
| GitHub Copilot | Not listed |
| Windsurf |