84 tagged eval, measured the same way as everything else here.
Browse within: harness-engineering 54sdd 54spec 54AI Safety 8prompt-engineering 8skill 5
Skill Claude CodeCodex
A Comet workflow step for turning an approved design into an implementation plan and completed code changes. Comet is the project workflow system described by the add-on.
Skill Claude CodeCodex
A Chinese-language workflow guide for starting or resuming a structured software change using OpenSpec and Superpowers.
Skill Claude CodeCodex
Comet Classic 阶段 1 —— 开启 OpenSpec change 并建立 proposal/design/tasks/.comet.yaml 产物。.
Skill Claude CodeCodex
Multi-persona idea evaluation with quality gate.
Skill Claude CodeCodex
Guidelines for using modern Angular APIs (signals, inject, control flow).
Skill Claude CodeCodex
This skill describes the mandatory standard operating procedure for using our internal SuperLint tool. Use this when tasks require fixing code quality issues according to corporate standards.
Skill Claude CodeCodex
Authors deterministic and LLM rubric graders for skillgrade evaluations. Use when creating scoring scripts, writing evaluation rubrics, or combining multiple graders with weighted scoring. Don't use for setting up eval pipelines, configuring eval.yaml defaults, or general test writing.
Skill Claude CodeCodex
Review a change in the EvalBench repo for (a) does it actually work — verified by running the tests and style checks, (b) does it follow EvalBench architecture — base-class contracts, config-key registration, PYTHONPATH-relative imports, sandbox isolation, concurrency safety, docs, (c) does it still build and deploy …
Skill Claude CodeCodex
Stage 3 of Clinical ASR Flywheel. Score a NeMo manifest, produce the five-section KER leaderboard (by-ipasource diagnostic). Not for ASR auth (/riva-asr).
Skill Claude CodeCodex
Improve AI application with evaluation-driven development. Define eval criteria, instrument the application, build golden datasets, observe and evaluate application runs, analyze results, and produce a concrete action plan for improvements. ALWAYS USE THIS SKILL when the user asks to set up QA, add tests, add evals…
Skill Claude CodeCodex
Evaluate AI agent output quality, safety, and cost using the Iris MCP server. Use when building, testing, or shipping agents and the user wants to score output quality, detect PII or prompt injection, verify citations, track cost per query, enforce cost budgets, add tracing/observability to an agent, or set up…
Skill Claude CodeCodex
Evaluate AI agent outputs for quality, safety, and cost using the Iris MCP server. Use when reviewing agent responses, checking for PII leaks, scoring output quality, or tracking execution costs.
Skill Claude CodeCodex
Use this skill when the user wants to commit, write a commit message, or save their staged or current changes with a clear message to git history.
Skill Claude CodeCodex
Use this skill when the user wants to draft, write, or generate a pull request (PR) description, summary, or overview for a branch, diff, or set of commits to prepare for code review.
Skill Claude CodeCodex
Use before launching any LLM feature or when monthly API costs are growing unexpectedly. Requires token count measurement, call volume analysis, and cost projection at 10× scale. Blocks "it's cheap enough now" completions.
Skill Claude CodeCodex
Use when prompt cost is too high, latency is above threshold, or context window limits are being approached. Requires measurement before and after each reduction. Blocks "I shortened the prompt so it should be cheaper" completions.
Skill Claude CodeCodex
Use before merging, deploying, or demo'ing any LLM feature. Requires documented eval results — pass rate, failure analysis, baseline comparison. Blocks "it looked good when I tested it" completions.
Skill Claude CodeCodex
Nothing is done until it has been run. Every claim that code works, a fix landed, or a result is correct must carry the command that proved it and that command's real output. Use whenever the user asks to implement, fix, change, refactor, debug, deploy, or verify code or config, or says "prove it", "did you test…
Skill Claude CodeCodex
Helps you write better code.
Skill Claude CodeCodex
Build, lint, and evaluate agent skills that demonstrably change behaviour, instead of writing adjectives that change nothing. Use whenever the user asks to create a skill, write a SKILL.md, improve or fix an existing skill, make a skill "actually work", or evaluate whether a skill does anything. Also use when the user…
Skill Claude CodeCodex
Before finalizing structured or extracted output in a domain with declared safety invariants, verify the output against your invariant suite and refuse to ship when a locked field was model-auto-filled. Invoke when the user or task involves high-stakes structured extraction, compliance fields, or pass/fail results…