Eval-Driven Development

Elevated · Review

Eval-driven development framework: define pass/fail criteria before implementation, capability/regression/consistency eval types, pass@k reliability metrics, grader patterns, and continuous eval integration during development.

evalseddmetricsreliabilitybenchmarking

Governance Receipt

Signer
sovereign-claw-ed25519
Signed At
6/4/2026
Risk Tier
T2
Receipt Hash
fba7606a
Manifest Hash
e1512015c2a055d59491179c65175bbb00a40d3ace8be795addbce07d312902f
Signature
5ONQLPxL
Root Public Key
349b0348

Skill Details

Gate Verdict
Elevated · Review
Publication State
published
Risk Tier
T2
Manifest Hash
e1512015

Download Attested Bundle

Claude Code SkillMCP ToolsRaw Bundle
Attested Source RepositoryOriginal Source

Install Eval-Driven Development

Each bundle is served from the attested copy of Eval-Driven Development, so the file you download is the one covered by the governance receipt on this page rather than whatever currently sits at the upstream source.

Claude Code skill

curl -L "https://atestiv.com/api/download/ecc-eval-driven-dev?format=claude-code" -o ecc-eval-driven-dev.skill.md

MCP tool definitions

curl -L "https://atestiv.com/api/download/ecc-eval-driven-dev?format=mcp" -o ecc-eval-driven-dev.mcp-tools.json

Raw attested bundle

curl -L "https://atestiv.com/api/download/ecc-eval-driven-dev?format=raw" -o ecc-eval-driven-dev.bundle.json

Verify the bundle against its manifest hash

shasum -a 256 ecc-eval-driven-dev.bundle.json

Provenance

  • Catalog identifier: ecc-eval-driven-dev
  • Harvested from the upstream repository.
  • The manifest hash below is what the signing key committed to. A bundle whose digest does not match it is not the attested Eval-Driven Development.
  • Anything published here has been through the same gate, so a missing receipt is itself a signal rather than an omission.

More Skills

Agent Benchmarking

Elevated · Review

Head-to-head coding agent comparison tool: YAML task definitions with judge criteria, git worktree isolation per agent run, pass rate/cost/time/consistency metrics, and reproducible benchmarking across Claude Code, Aider, Codex, and other agents.

Exa Neural Search

Elevated · Review

Neural search via Exa MCP: web search, code examples, company intel, people lookup, and AI-powered deep research with semantic retrieval.

fal.ai Media Generation

Elevated · Review

Unified AI media generation via fal.ai MCP: text-to-image, text/image-to-video with Seedance/Kling/Veo, text-to-speech with CSM-1B, and video-to-audio with ThinkSound.