skill-up vs SkillRank

Agent skill evaluation tools compared

skill-up and SkillRank both help developers evaluate skills for AI agents. skill-up focuses on repeatable test cases, workspace tasks, and reports, while SkillRank combines paired skill evaluations with a registry for finding and installing skills.

skill-upSkillRank
LanguageGoRust
LicenseApache-2.0MIT
Stars1.4k0
Forks1220
Last analyzedNov 2, 2026Oct 22, 2026

Key differences

  • skill-up uses YAML evaluation cases to test agent responses and workspace changes; SkillRank runs paired trials on a pinned fixture repository and reports task success, token, turn, time, and cost changes.
  • skill-up includes built-in engines such as Claude Code, Codex, and Qoder CLI, plus custom engine configuration; SkillRank supports skills for Claude Code, Codex, Copilot, and other agents, and exposes tools to Claude Code and Codex through MCP.
  • skill-up is a Go CLI with evaluation workflows for local development and CI; SkillRank is a Rust CLI and public registry with search, install, recommendations, and local registry options.
  • skill-up can output JSON, JUnit XML, HTML, and benchmark reports, and provides a GitHub Action; SkillRank displays published evaluation results in official, community-reported, and self-reported trust tiers.
  • skill-up is licensed under Apache-2.0; SkillRank's CLI and core library use MIT, while its registry service and website use Elastic License 2.0.
  • Both projects are recent and changing quickly. SkillRank's supplied considerations flag that no tests are detected despite README documentation of `cargo test`; skill-up's considerations note its Go 1.25 build requirement and Node.js runtime requirement.

Choose skill-up if you…

  • need repeatable YAML-based evaluations of agent responses and workspace changes.
  • want CI-oriented reports or GitHub Actions for evaluating skills and coding-agent tasks.
  • want to compare supported agent engines using the same case suite.
Read the skill-up analysis →

Choose SkillRank if you…

  • want to discover, install, or manage skills through a registry or local catalog.
  • need paired trials that report changes in task success, tokens, turns, time, and cost.
  • want Claude Code or Codex to search and manage skills through MCP tools.
Read the SkillRank analysis →

Compare the numbers side by side, or add more projects

This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.