by skillberry-ai · MCP Server · ★ 56
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
Optimize any AI agent’s skills, tools/MCP, and prompts against your own evals.
| Stars | 56 |
| Forks | 16 |
| Language | Python |
| Category | MCP Server |
| License | Apache-2.0 |
| Quality Score | 55.0341650787553/100 |
| Open Issues | 38 |
| Last Updated | 2026-09-15 |
| Created | 2026-06-15 |
| Platforms | claude-code, mcp, python |
| Est. Tokens | ~13k |
Looking for a cap-evolve alternative? If you're comparing cap-evolve with other mcp server tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Turn feature specs into merged PRs with a self-supervising swarm of coding agents — parallel execution, isolat
Automated loop driver, slash commands, council automation, MCP browser bridge, and portfolio governance for Cl
A collection of AI agent skills for Clawdbot, Claude Code, Codex
Hands-on tutorials for building AI agents from scratch. Learn LLM APIs, prompt engineering, tool calling, and
HeyClaude (formerly Claude Pro Directory) is a searchable collection of pre-built AI skills, agents, MCP serve
Explore other popular mcp server tools:
cap-evolve is Optimize any AI agent’s skills, tools/MCP, and prompts against your own evals.. It is categorized as a MCP Server with 56 GitHub stars.
cap-evolve is primarily written in Python. It covers topics such as agent-optimization, agent-skills, agentic-ai.
You can find installation instructions and usage details in the cap-evolve GitHub repository at github.com/skillberry-ai/cap-evolve. The project has 56 stars and 16 forks, indicating an active community.
cap-evolve is released under the Apache-2.0 license, making it free to use and modify according to the license terms.
The top alternatives to cap-evolve on Agent Skills Hub include OmoiOS, intellegix-code-agent-toolkit, agent-skills. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: