evaluator-dspy
Isolated DSPy optimizer companion (GEPA and MIPROv2) for governed self-evolution.
Skill by nficano
Chalk Labs
Need help deploying evaluator-dspy?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/nficano-evaluator-dspy
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/evaluator-dspy
What's inside
- Skill name
evaluator-dspy- Description
- Isolated DSPy optimizer companion (GEPA and MIPROv2) for governed self-evolution. Use when optimizing a text or prompt target against a sealed training and validation view.
- License
- MIT
- Instructions
- 19 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (4)
- SKILL.md
- manifest.yaml
- index.ts · src/
- types.ts · src/
SKILL.md (first lines)
# DSPy text optimizer Isolated companion adapter for GEPA and MIPROv2. Elliott supplies a sealed training/validation view, a materialized text target, limits, and a seed. The companion returns schema-validated candidate lineage only. It cannot access holdout data, Proposals, active configuration, Snapshots, Git credentials, or release operations. GEPA is the primary reflective optimizer. MIPROv2 is the policy-selected fallback. Model calls are made through Elliott-authorized routes rather than ambient companion credentials. The image validates the train/validation-only wire view, enforces run budgets, and uses a loopback model proxy with a short-lived scoped token. Optimization jobs are real process groups with bounded time slices, pause, resume, and idempotent cancellation. Fixture mode is limited to build and protocol smoke tests.
From nficano/elliott, MIT.
More from nficano/elliott
- 0.547score
evaluator-agent-benchmarks
nficano
Isolated benchmark runner and independent evolution evaluator that owns the hidden comparison route and broad benchmark ladder.
SkillBFreshness: B★ 1Claude Code +9 - 0.547score
deep-trace
nficano
Read-only observability map of the running agent — live topology, sessions, telemetry, and container health.
SkillBFreshness: B★ 1Claude Code +9 - 0.542score
mcp-client
nficano
Connect to Streamable HTTP and SSE Model Context Protocol servers and expose their tools, resources, and prompts.
SkillBFreshness: B★ 1Claude Code +9 - 0.542score
evaluator-darwinian
nficano
Isolated Darwinian Evolver companion for governed self-evolution.
SkillBFreshness: B★ 1Claude Code +9 - 0.542score
scheduler
nficano
Run due work in fresh frames under a lease, re-resolving authority at fire time.
SkillBFreshness: B★ 1Claude Code +9 - 0.541score
cloudflared
nficano
Provision and maintain the Cloudflare tunnel that carries inbound webhooks, and watch that it stays up.
SkillBFreshness: B★ 1Claude Code +9 - 0.541score
glitchtip
nficano
Ship runtime errors to a Sentry-compatible collector (GlitchTip or Sentry), on by default with zero setup.
SkillBFreshness: B★ 1Claude Code +9 - 0.541score
gateway-slack
nficano
Chat with the agent in Slack via Socket Mode and the agent view.
SkillBFreshness: B★ 1Claude Code +9 - 0.538score
gateway-bluebubbles
nficano
Bridge iMessage through a paired BlueBubbles server.
SkillBFreshness: B★ 1Claude Code +9 - 0.538score
gateway-gmail
nficano
Search, read, and reply to Gmail with OAuth access tokens.
SkillBFreshness: B★ 1Claude Code +9 - 0.536score
traefik
nficano
Publish LAN services through Traefik with automatic HTTPS certs.
SkillBFreshness: B★ 1Claude Code +9 - 0.536score
gateway-webhook
nficano
Accept inbound webhooks only on declared, HMAC-verified routes.
SkillBFreshness: B★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Aug 12, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/nficano/elliott
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from nficano
- 0.547score
deep-trace
nficano
Read-only observability map of the running agent — live topology, sessions, telemetry, and container health.
SkillBFreshness: B★ 1Claude Code +9 - 0.547score
evaluator-agent-benchmarks
nficano
Isolated benchmark runner and independent evolution evaluator that owns the hidden comparison route and broad benchmark ladder.
SkillBFreshness: B★ 1Claude Code +9 - 0.542score
scheduler
nficano
Run due work in fresh frames under a lease, re-resolving authority at fire time.
SkillBFreshness: B★ 1Claude Code +9 - 0.542score
evaluator-darwinian
nficano
Isolated Darwinian Evolver companion for governed self-evolution.
SkillBFreshness: B★ 1Claude Code +9 - 0.542score
mcp-client
nficano
Connect to Streamable HTTP and SSE Model Context Protocol servers and expose their tools, resources, and prompts.
SkillBFreshness: B★ 1Claude Code +9 - 0.541score
glitchtip
nficano
Ship runtime errors to a Sentry-compatible collector (GlitchTip or Sentry), on by default with zero setup.
SkillBFreshness: B★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get evaluator-dspy running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
evaluator-dspy skill. Agent Mart. https://agentmart.dev/skills/nficano-evaluator-dspy (last updated 2026-10-24)