skillberry-ai

37 mods across 2 repositories, 52 stars between them.

finalize

25

skillberry-ai/cap-evolve

Skill Claude CodeCodex

Score the best candidate on the held-out TEST split exactly once and seal the run. Use as the last evaluation step, after optimization stops. The run dir enforces the seal — a second finalize raises an error — so the headline number is produced once on data the optimizer never saw, the way an honest benchmark result…

47 yesterday A 69 tokens original Apache-2.0

gate

26

skillberry-ai/cap-evolve

Skill Claude CodeCodex

Apply the acceptance decision that keeps optimization honest — always on the val split, by default requiring the improvement to exceed the significance bar (Δ > k·SE) so noise is not mistaken for progress. Use to inspect or reproduce a single accept/reject decision; the algorithms apply it internally every iteration.

47 yesterday A 62 tokens original Apache-2.0

implement-and-check

27

skillberry-ai/cap-evolve

Skill Claude CodeCodex

Runs the hard gate that has to pass before any optimization budget is spent. Use right after intake. Walks the agent through implementing the 3 required adapter methods plus any defaulted hooks that need overriding (and any selected skill's abstract methods), then runs cap-evolve check on the project plus each…

47 yesterday A 94 tokens original Apache-2.0

intake

28

skillberry-ai/cap-evolve

Skill Claude CodeCodex

Starts a cap-evolve optimization run. Interviews the user to decide what capability to optimize, which runner/optimizer/algorithm to use, and where the tasks and the scoring source live, then scaffolds .capevolve/project/ (adapter stub, capevolve.yaml, PROJECT.md). Use when someone asks to optimize or improve an agent…

47 yesterday A 169 tokens original Apache-2.0

report

29

skillberry-ai/cap-evolve

Skill Claude CodeCodex

Summarize a run for a human — baseline val → best val → sealed test, the winning candidate, iterations spent, and pass^k. Use after finalize. Writes report.md and prints a compact JSON summary; the source of truth for "did this optimization actually work, and by how much".

47 yesterday A 63 tokens original Apache-2.0

<skill-name>

30

skillberry-ai/cap-evolve

Skill Claude CodeCodex

Concrete triggers. What situation in the pipeline calls for it.

47 yesterday A 50 tokens original Apache-2.0

mcp-builder

31

skillberry-ai/runspace-agent

Skill Claude CodeCodex

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

5 1mo ago A 61 tokens copy · 100% Apache-2.0

skill-creator

32

skillberry-ai/runspace-agent

Skill Claude CodeCodex

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

5 1mo ago A 64 tokens copy · 88% Apache-2.0

analyzer

33

skillberry-ai/runspace-agent

Agent Claude Code

Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.

5 1mo ago A 0 tokens copy · 100% Apache-2.0

grader

35

skillberry-ai/runspace-agent

Agent Claude Code

Evaluate expectations against an execution transcript and outputs.

5 1mo ago A 0 tokens copy · 100% Apache-2.0