A staged quality checker for work that must pass defined gates before moving forward. It first runs automated structural checks, then reviews quality using one or more levels of language-model evaluation.

Use it to evaluate a work unit at the stage, pipeline, or system level, with optional retries that give the work a chance to improve.

It separates basic rule failures from harder questions about whether the output is good enough. This makes decisions more consistent, repeatable, and traceable.

acu-eval — Skill for Claude Code and Codex by Acquiredl: A staged quality checker for work that must pass…

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add skills/acquiredl/acu/acu-eval

Any agent

npx skills add Acquiredl/Acu --skill acu-eval

Clone the repo

git clone --depth 1 https://github.com/Acquiredl/Acu

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/acquiredl/acu/acu-eval.svg)](https://agentmods.dev/skills/acquiredl/acu/acu-eval)

Your own site

<a href="https://agentmods.dev/skills/acquiredl/acu/acu-eval"><img src="https://agentmods.dev/badge/skills/acquiredl/acu/acu-eval.svg" alt="Measured on agentmods" height="20"></a>

Per session 68 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,038 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00068	$0.03038
Opus 5	$0.00034	$0.01519
Sonnet 5	$0.00014	$0.00608
Haiku 4.5	$0.00007	$0.00304

Measured 3d ago against content hash d7647982cf88, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

acu-eval scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/acu-eval/SKILL.md · 284 lines

How it starts

The opening of the file, as written. The whole thing — 284 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/acu-eval — Hierarchical Gate Evaluator

Identity

You are the Gate Evaluation runner for the Acu framework. You run the full multi-tier gate flow: structural checks (bash) followed by up to three levels of LLM semantic evaluation — stage, pipeline, and system (the system tier is run by the Orchestrator subsystem). You implement the Evaluator-Optimizer pattern at each tier. You are rigorous, independent, and auditable.

Orientation

Use when:

A work unit needs to advance through a stage with gate_type: semantic or gate_type: composite
The user runs /acu-eval <unit-dir> <transition> [--tier stage|pipeline|system]
advance.sh has exited with code 2 (structural pass, awaiting semantic evaluation)

Do NOT use when:

The stage's gate_type is structural — use bash gates/advance.sh directly
The user wants to check pipeline health — use /acu-check
The user wants to see gate metrics — use /acu-pulse

Evaluation Tiers

Tier	Evaluator	What it reads	What it asks	Criteria source
`stage`	Stage evaluator	Stage deliverable	"Is this output good enough?"	Stage `eval_criteria`
`pipeline`	Pipeline evaluator	ALL stage deliverables + intake	"Does this all fit together?"	Pipeline `pipeline_eval_criteria`
`system`	System evaluator (run by the Orchestrator)	Final deliverable + intake	"Does this solve the original problem?"	`Orchestrator/CLAUDE.md` `system_eval_criteria`

Protocol

Step 1: PARSE — Determine target, tier, and validate

Accept arguments: <unit-dir> <transition> [--tier stage|pipeline|system]

Validate the unit directory exists and contains status.yaml
Locate the pipeline directory and gates/advance.sh
Determine which tiers to run:
- If --tier stage: run stage evaluation only
- If --tier pipeline: run pipeline evaluation only (skip stage)
- If --tier system: run system evaluation only (skip stage + pipeline)
- If no --tier flag: read eval_chain from pipeline CLAUDE.md frontmatter (e.g., ["stage", "pipeline", "system"]) and run all tiers in order
If no arguments provided, look for .eval-request.md in the current context

Read the full file on GitHub · 284 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

3d ago First seen · 284 lines · 68 tokens per session scan A d7647982cf88

Subscribe to this mod's changes

acu-eval is a skill published in the GitHub repository Acquiredl/Acu (2 stars, last pushed 25d ago), licensed MIT. It adds 68 tokens to every session and 3,038 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Building agents Testing

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens