Run Evals

Eval-Driven Development — define capability and regression evals before building; code graders use verify commands, model graders use explicit rubrics; log pass@k. Use before develop-tdd on new features, or when measuring agent capability over runs.

292 stars 23 forksAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add danielvm-git/bigpowers --skill run-evals
Manual install
git clone --depth 1 https://github.com/danielvm-git/bigpowers.git /tmp/bigpowers
mkdir -p ~/.claude/skills
cp -r /tmp/bigpowers/.cline/skills/run-evals ~/.claude/skills/run-evals

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Run Evals

> **HARD GATE** — Define evals before implementation. Code graders = runnable `verify:` commands; model graders = explicit rubric with pass/fail criteria.

## Process

1. Name the capability under test (one sentence).
2. Write `specs/EVALS-<feature>.md` with:
   - **Capability evals** (does it do the job?)
   - **Regression evals** (did we break anything?)
3. Assign grader type per eval: `code` (shell verify) or `model` (rubric).
4. Assign **strictness tier** per eval (graduated promotion — e45s37):

   | Tier | Meaning | Promotion rule |
   |------|---------|------------------|
   | `EXPERIMENTAL` | New eval, may flake | Not gating |
   | `USUALLY_PASSES` | Stable in dev; ≥2/3 recent runs pass | Blocks BUILD only when combined with ALWAYS_PASSES suite |
   | `ALWAYS_PASSES` | Zero tolerance; required for release | Any single failure blocks BUILD and merge |

   Promote: `EXPERIMENTAL → USUALLY_PASSES` after 3 consecutive passes; `USUALLY_PASSES → ALWAYS_PASSES` after 5 consecutive passes with zero flakes documented in `specs/state.yaml`.

5. Run evals; log results table with pass@k (e.g. 3/3 runs) and tier per eval.
6. Block BUILD phase until all `ALWAYS_PASSES` evals pass at agreed k. `USUALLY_PASSES` failures warn; `EXPERIMENTAL` failures log only.

## Artefact
Read the full skill on GitHub

Details

Skill name
run-evals
Path
.cline/skills/run-evals/SKILL.md
License
MIT
Repo updated
9 November 2026

FAQ

What does the Run Evals skill do?

Eval-Driven Development — define capability and regression evals before building; code graders use verify commands, model graders use explicit rubrics; log pass@k. Use before develop-tdd on new features, or when measuring agent capability over runs.

How do I install the Run Evals skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add danielvm-git/bigpowers --skill run-evals. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Run Evals popular?

It lives in danielvm-git/bigpowers, which has 292 GitHub stars and 23 forks.

Related skills

Align Grid

292

danielvm-git/bigpowers

Build editorial/magazine/report webpages on a GENUINE Müller-Brockmann modular grid (International Typographic Style) — not a decorative one. Encodes the discipline (columns + modules + baseline, grotesque type, flush-left, restrained black/white/red palette) AND the hard-won front-end engineering to make the grid real, visible, and verified: one CSS-variable source of truth, an interactive grid-toggle overlay that lives in the SAME content box as the content, subgrid "bands" so every element snaps to a column line, an 8px baseline lock, and runtime OPTICAL ALIGNMENT that puts display type's i

Writing & Content23

Assess Impact

292

danielvm-git/bigpowers

Analyze the blast radius of a proposed change before any code is written. Maps dependents, affected stories, and test coverage. Produces specs/IMPACT_LATEST.md. Use before plan-work on any non-trivial change, when touching a shared module, or when the user asks "what does this break?".

Testing & QA23

Audit Code

292

danielvm-git/bigpowers

Self-review checklist for the coding agent to run before dispatching a reviewer. Checks CONVENTIONS.md compliance, Boy Scout Rule, test coverage, types, and SOLID. Produces a pass/fail checklist. Use before request-review, before committing, or when user asks for a code quality check.

Software Development23

Seed Conventions

292

danielvm-git/bigpowers

Generate CLAUDE.md and CONVENTIONS.md for a brand-new project through a brief interview, and create the specs/ directory with evolved bigpowers structure (product/, tech-architecture/, verifications/, epics/archive/). Entry point for greenfield projects. Use when starting a new project from scratch, when user asks to set up AI agent conventions, or when there is no CLAUDE.md yet.

AI, LLMs & Agents23

Reset Baseline

292

danielvm-git/bigpowers

Restore the project to a known clean state between agent runs or experiments. Use between benchmark runs, after a failed spike, or when user wants a clean working tree.

AI, LLMs & Agents23

Simulate Agents

292

danielvm-git/bigpowers

Run Mock User and Auditor agents against a feature in fresh contexts before human review. Use after verify-work, before request-review, when user wants pre-review simulation.

AI, LLMs & Agents23

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.