Agent Skills

Create Harbor Task

harbor

Scaffold, configure, and verify a Harbor evaluation task end-to-end, including prompts, environment, verifier selection, and Oracle solution.

Prompt Engineering

ai-driven-development

Techniques and templates for crafting precise, reliable prompts and agentic instructions across LLMs.

Arize Experiment

awesome-copilot

Create, run, and analyze Arize experiments to evaluate and compare model performance using the ax CLI.

Semantic Gradient Optimization (SGO)

sgo

Optimize a product, resume, pitch, or profile against simulated evaluator populations using LLM scoring and counterfactual probes to find changes that improve r

OmniDocBench Evaluation Helper

opendatalab

Run, validate, and parse OmniDocBench document parsing evaluations with Docker/conda workflows and result parsing.

Skill Creator (Claw)

skill-creator-claw

Create, test, and iteratively improve OpenClaw skills; includes eval workflows, test-case guidance, and packaging tools.

LLM Council — Multi-model Consensus Dashboard

happycapy-skills

Run parallel queries against multiple models and view side-by-side responses in a live dashboard; synthesize consensus and run anonymous model voting.

Skill Creator

my_arxiv_daily

Create, iterate on, and evaluate Agent Skills: draft SKILL.md, run test prompts, collect quantitative and qualitative feedback, and optimize skill descriptions

Brain in the Fish — Document Evaluation

brain-in-the-fish

Evaluates documents against criteria using multi-agent, ontology-grounded scoring to produce evidence-backed, auditable evaluation reports.

Calibrate — Agent & Skill Benchmarking

ai-rig

Runs synthetic benchmarks and calibration tests for agents and skills: measures recall, precision, confidence calibration, and A/B comparisons to quantify instr

Skill Creator

awesome-omni-skill

Guided workflow for drafting, testing, and iterating Agent Skills: write SKILL.md, run evals, grade outputs, and improve descriptions to improve triggering accu

Skill Creator

claude-superskills

Create, improve, and evaluate Agent Skills with a guided workflow: capture intent, draft SKILL.md, run evals and benchmarks, and optimize triggering description

Scout

DeepScientist

Frame ambiguous research or engineering tasks: clarify goals, metrics, datasets, and a justified baseline so work can proceed with confidence.

LLM Evaluation

claude-plugins

Evaluation framework and tools for systematically measuring LLM performance using automated metrics, human judgment, and A/B testing.

RePrompter

reprompter

Turn messy prompts into structured, scoreable prompts or multi-agent plans; also reverse-engineer great outputs into reusable prompts.

Cortex — ML Pipeline Builder

tonone

Build a reproducible ML training-to-serving pipeline: data validation, feature engineering, training, evaluation, and a serving endpoint.

ISC-Bench

ISC-Bench

Benchmark and pipeline for evaluating model safety (Internal Safety Collapse) using the TVD framework; builds prompts, runs models, extracts harmful content, an

Code Evaluation Harness

my-cc-harness

Spawns an independent evaluator agent to score code outputs on functionality, code quality, originality, and usability/security, producing a structured EVAL_REP

Harness Health Audit

my-cc-harness

Assess and score the overall health of a Claude Code harness across architecture, skills coverage, hooks, rules, MCP servers, eval pipelines, and team setup.

Hugging Face Evaluation Manager

skillattack

Extract, import, and add structured model evaluation results to Hugging Face model cards; run or import benchmark evaluations and generate model-index YAML for

Review Merge Worker

Claude Code Skills

Merges and deduplicates evaluation worker results into a single verified review summary for coordinators, applying an architecture gate and verification checks.

Building Agent Systems

code-abyss

Reference guide for designing single- and multi-agent LLM systems, RAG pipelines, security hardening, and prompt engineering for production agents.

Langium-AI (LAI)

langium-ai

CLI to generate language descriptors, synthesize system prompts, and run evaluations to iteratively refine AI tooling for Langium projects.

OpenSearch Launchpad

opensearch-agent-skills

Guides users to design, build, and run OpenSearch-based search applications — from data ingestion and index design to semantic/hybrid search and evaluation.

OpenJudge — LLM Evaluation Pipeline

openjudge

Tools and patterns to build automated evaluation pipelines for LLMs: graders, runners, aggregators, and analysis utilities for comparing model outputs and scori

Skill Creator

sage

Create, iterate on, and evaluate Agent Skills: write SKILL.md, design test cases, run benchmarks, and improve triggering and performance.

Claw Bench

claw-bench

Benchmarking skill that guides an agent through a structured suite of capability tests and reporting steps for leaderboard submission.

Evaluation Filtering

de-anthropocentric-research-engine

Multi-stage evaluation pipeline that scores ideas for novelty, assesses feasibility, ranks them, and selects top candidates for further development.

Nemotron 3 Nano Reference

nemotron

Authoritative knowledge-base for Nemotron 3 Nano: architecture, training data, SFT/RL recipes, evaluation and deployment guidance.

NeMo Evaluator

nemo-platform

Turn a domain rubric and dataset into a reproducible evaluation using the NeMo Evaluator SDK; generate configs, run local evaluations, and explain scores and fa