Tagged evals

Create, edit and benchmark agent skills with evals and variance analysis

Score a local skill or plugin, measure token use and benchmark starter scenarios

Find command injection, eval misuse and arbitrary code loading in agent systems

Audit an agent harness for broken paths, redundant rules and unused skills

Grade how completely an implementation fulfills a PRD case by case

Compile files and folders into Harbor tasks and run map-reduce eval jobs

Audit Claude Code and Codex subagent files for loadability, scope and overlap

Audit Claude Code and Codex hooks for validity, safety and exit handling

Audit Claude Code rules files for broken globs, bad frontmatter and noisy rules

Audit agent skills for metadata, trigger accuracy and completion criteria before release

Run and analyze evaluations for ADK agents with eval datasets and failure review

Scaffold eval specs with stimuli, graders and rubrics for agent skills and workflows

Find why MSBuild project evaluation is slow, such as huge globs and deep imports

Build durable backend AI agents with tools, skills, schedules and evals

Run one headless voice pack reroll job for the peon eval pipeline

Create, evaluate, package and upload agent skills with eval runs

Create, improve and benchmark agent skills with evals and description tuning

Evaluate LLM outputs with judge models, rubrics and pairwise comparisons

Score clinical ASR manifests and produce a KER leaderboard by ASR source

Create and modify NeMo AutoModel training and eval recipes in YAML

Integrate the NeMo Fabric Python SDK into an app, service or eval harness

Score RAG answer quality with RAGAS on a corpus and training questions

Run a Python training or eval script in an existing virtualenv, no containers

Judge whether a slow code path is a good candidate to port to NVIDIA Warp

Audit eval coverage for an executed AI phase and list remediation steps

Score RAG retrieval and citation quality from JSONL fixtures in offline evals

Write and test GoClaw agent skills with evals, benchmarks and description tuning

Design metrics, calibration and eval suites that show whether an agent really improved

Review finished work and distill lessons into reusable methods, skills and evals

Create, improve and benchmark agent skills with evals and description tuning

Evaluate Expo skills for trigger accuracy, code quality and simulator screenshots

Add evaluators, online monitoring and quality gates to measure and speed up an agent

Diagnose one agent run step by step in an incident-style canvas report

Create and run evaluation suites for M365 Copilot declarative agents

Create, improve and evaluate agent skills from workflows, SOPs and scripts

Generate eval tests for any MCP server with Jest or Vitest using mcpjam SDK

Turn MCPJam Explore test cases into one SDK eval test per case

Convert promptfoo, pytest or Jest test suites into an MCPJam eval file

Launch MCPJam hosted evals, poll them to a verdict and triage failing steps

Manage Rightbrain tasks, agents, approvals and evals with a local mirror

Set up an eval harness to score Claude Code sessions with eval-driven development

Audit an LLM eval pipeline for missing error analysis and unvalidated judges

Turn student course evaluations into a prioritized teaching improvement plan

Turn eval study numbers into on-brand charts and check them with Playwright

Find the right evals starting point and route to the matching evaluation skill

Write code evaluators that check objective failure rules against agent traces

Benchmark a third-party LLM endpoint for availability, speed, fidelity and quality

Configure Promptfoo evals with LLM-as-judge rubrics and custom Python assertions

Create, edit and benchmark agent skills, then run evals to measure them