eval-harness

eval-harness is a skill for Claude Code, Codex from majiayu000/vibeguard. It costs 29 tokens per session (909 once invoked), scanned A, original, MIT.

An evaluation workflow for measuring the quality and reliability of code generated by an agent. pass@k measures whether at least one of several attempts succeeds, while pass^k measures whether all attempts succeed.

In plain words
What is it for?
Use it when comparing prompts, skills, hooks, or workflows; building regression gates; or turning review criteria into deterministic or model-graded scores.
Why use it?
It replaces relying on a single successful example with repeatable checks for improvements, repairs, regressions, and unrelated tasks.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it when comparing prompts, skills, hooks, or workflows; building regression gates; or turning review criteria into deterministic or model-graded scores.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/majiayu000/vibeguard/eval-harness
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.

Any agent
npx skills add majiayu000/vibeguard --skill eval-harness
Clone the repo
git clone --depth 1 https://github.com/majiayu000/vibeguard

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.

agentmods badge for eval-harness

README.md
[![agentmods](https://agentmods.dev/badge/skills/majiayu000/vibeguard/eval-harness/github.svg)](https://agentmods.dev/skills/majiayu000/vibeguard/eval-harness)
Your own site
<a href="https://agentmods.dev/skills/majiayu000/vibeguard/eval-harness"><img src="https://agentmods.dev/badge/skills/majiayu000/vibeguard/eval-harness/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for eval-harness

Your own site · 80×15
<a href="https://agentmods.dev/skills/majiayu000/vibeguard/eval-harness"><img src="https://agentmods.dev/badge/skills/majiayu000/vibeguard/eval-harness.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 29 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 909 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00029 $0.00909
Opus 5 $0.00015 $0.00454
Sonnet 5 $0.00006 $0.00182
Haiku 4.5 $0.00003 $0.00091

Measured 10d ago against content hash 9b18e95c1566, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

eval-harness 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 10d 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.

skills/eval-harness/SKILL.md · 105 lines

How it starts

The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Eval Harness

Overview

Evaluation-driven development: Not just "can the code run", but quantify "how good is the code".

When to Activate

  • Designing or changing an evaluation harness for agent behavior or guard quality.
  • Comparing prompt, skill, hook, or workflow variants with measurable outcomes.
  • Adding a regression gate that must prove repair without hiding unrelated-task regressions.
  • Converting subjective review criteria into deterministic or model-graded metrics.
  • A guard, hook, workflow, or skill needs measurable repair/regression evidence before adoption.
  • A user asks whether an agent workflow is actually improving quality rather than only passing one example.
  • A change affects scoring, grading, or benchmark thresholds.

Red Flags

  • Only one hand-picked success case exists and there is no held-out or unrelated task coverage.
  • A probabilistic grader is used where a deterministic build, test, lint, or coverage check is available.
  • pass@k improves while pass^k or unrelated-task behavior regresses.

Checklist

  • Define deterministic checks before adding model-judged grading.
  • Record target and unrelated scenarios with before/after outcomes.
  • Report pass@k, pass^k, and regression counts separately.

Core indicators

pass@k (single success rate)

  • Generate k candidate solutions, with a probability of at least 1 passing
  • Used to evaluate the completion quality of a single task
  • Target: pass@1 > 80%

pass^k (continuous success rate)

  • The probability of passing all k consecutive tasks at once
  • Used to evaluate overall workflow reliability
  • Goal: pass^5 > 50% (pass all 5 consecutive tasks in one go)

Grader type

Code Basics Grader (deterministic)

Grader Check content Pass conditions
Compilation check Whether the code can be compiled / type check passed Zero errors
Test check Whether all tests passed Full green
Lint check Whether the code style conforms to the specification Zero warnings (or only allowed warnings)
Coverage check Check whether the test coverage reaches the standard ≥ 80%

Read the full file on GitHub · 105 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.

  1. 10d ago First seen · 105 lines · 29 tokens per session scan A 9b18e95c1566

Subscribe to this mod's changes

eval-harness is a skill published in the GitHub repository majiayu000/vibeguard (41 stars, last pushed yesterday), licensed MIT. It adds 29 tokens to every session and 909 once invoked, about $0.0001 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 2026-08-30.

Related

Other skills, from other repositories

setup

Set up or update TDD Guard for the current project. Detects the test framework, installs or updates the matching reporter, and configures or migrates its configuration to match the current specification.

nizos/tdd-guard · 39 tokens

coding-agents-hooks-authoring

To author, register, and test Rosetta hooks, add a SemanticKind, or debug a hook that won't fire.

griddynamics/rosetta · 31 tokens

ring:running-dev-cycle

Running the backend dev cycle: implements every task in a rolling-wave plan.md (ring:writing-plans format) for a Go/TS service, driving specialist agents through Gate 0 implementation/TDD, Gate 8 parallel review, and Gate 9 validation per epic, elaborating later phases at each phase boundary. Use when starting or…

LerianStudio/ring · 122 tokens

ring:instrumenting-streaming-events

Instrumenting streaming events: wires lib-streaming event emission end-to-end into a Lerian Go service via a 13-gate cycle (catalog, Builder bootstrap, Emit sites, outbox, HTTP manifest, NoopEmitter fallback, integration and chaos tests), dispatching ring:backend-go under TDD. Consumes the validated…

LerianStudio/ring · 102 tokens

ring:implementing-tasks

Implementing a single planned task (Task N.M.T) end-to-end: selects the right backend agent by language and service type, drives one TDD RED->GREEN turn, then verifies coverage, lint, license headers, runtime, and delivery before handoff. Runs as Gate 0 before ring:reviewing-code. Use to drive ONE task inside an…

LerianStudio/ring · 103 tokens

ring:writing-skills

Writing or editing a Ring skill: SKILL.md structure, frontmatter and Agent-Search-Optimization rules, token-efficiency targets, and bulletproofing (Iron Law, rationalization tables, Red Flags) so discipline-enforcing skills resist excuses. Use when creating or revising a skill. Delegates pressure-testing to…

LerianStudio/ring · 100 tokens