verify-serving-parity

verify-serving-parity is a skill for Claude Code, Codex from understudylabs/understudy-agent-tools. It costs 24 tokens per session (366 once invoked), scanned A, original, MIT.

An offline checker for determining whether two or more model-serving lanes produce equivalent results from the same approved local artifacts. Serving lanes are separate systems or configurations that run a model.

In plain words
What is it for?
Use it to compare serving lanes with paired synthetic samples, render and contract fingerprints, parser evidence, artifact references, and SHA-256 hashes.
Why use it?
It prevents comparisons based on missing samples, mismatched model artifacts, invalid hashes, or weak evidence, while keeping the evidence status visible.

Skill for Claude CodeCodex

Part of the understudy plugin — 43 skills, 1 command shipped together

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/understudylabs/understudy-agent-tools/verify-serving-parity
Any agent
npx skills add understudylabs/understudy-agent-tools --skill verify-serving-parity
Clone the repo
git clone --depth 1 https://github.com/understudylabs/understudy-agent-tools

Made for: Claude Code, Codex.

Or install understudy, the plugin that ships this one along with the rest of its 43 skills, 1 command.

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 verify-serving-parity

README.md
[![agentmods](https://agentmods.dev/badge/skills/understudylabs/understudy-agent-tools/verify-serving-parity.svg)](https://agentmods.dev/skills/understudylabs/understudy-agent-tools/verify-serving-parity)
Your own site
<a href="https://agentmods.dev/skills/understudylabs/understudy-agent-tools/verify-serving-parity"><img src="https://agentmods.dev/badge/skills/understudylabs/understudy-agent-tools/verify-serving-parity.svg" alt="Measured on agentmods" height="20"></a>
Per session 24 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 366 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.

ModelPer sessionOnce invoked
Fable 5 $0.00024 $0.00366
Opus 5 $0.00012 $0.00183
Sonnet 5 $0.00005 $0.00073
Haiku 4.5 $0.00002 $0.00037

Measured 3d ago against content hash 62ff5b594953, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

verify-serving-parity 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.

skills/verify-serving-parity/SKILL.md · 39 lines

What it actually says

Verify serving parity

Use src/serving-parity as a pure, offline verifier. Every lane must provide an observed render fingerprint, contract fingerprint, parser evidence, and an artifact reference plus SHA-256. Outputs contain references and hashes only; never place prompts, responses, traces, labels, credentials, or weights in a parity artifact.

Declare a meaningful minimum paired sample before scoring (default: 20). A failed preflight refuses comparison. observed, weak, deviation, and failed evidence status must remain visible, and the output includes understudy.promotion_receipt.v1 compatibility metadata for downstream gates.

Run only against synthetic or already-approved local artifacts; this verifier makes no provider calls and does not access holdout data.

Safety Gates

  • Compare at least two uniquely named lanes using the same artifact hash, observed render fingerprint, contract fingerprint, protocol, sampling, stop sequences, parser contract, and frozen task IDs.
  • Predeclare a paired-sample floor and equivalence band. Missing pairs, duplicate task IDs, deviations, weak evidence, or invalid hashes fail closed.
  • Keep prompts, responses, labels, traces, credentials, and weights out of the parity artifact. A passing parity receipt is necessary but not sufficient for promotion.

Resolve CLI

There is no public CLI yet. Build the package and call preflightServingParity followed by scoreServingParity from dist/serving-parity/index.js in an offline script. Do not score when preflight refuses the evidence.

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. 3d ago First seen · 39 lines · 24 tokens per session scan A 62ff5b594953

Subscribe to this mod's changes

verify-serving-parity is a skill published in the GitHub repository understudylabs/understudy-agent-tools (16 stars, last pushed 4d ago), licensed MIT. It adds 24 tokens to every session and 366 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

instrument-data-to-allotrope

Convert laboratory instrument output files (PDF, CSV, Excel, TXT) to Allotrope Simple Model (ASM) JSON format or flattened 2D CSV. Use this skill when scientists need to standardize instrument data for LIMS systems, data lakes, or downstream analysis. Supports auto-detection of instrument types. Outputs include full…

anthropics/knowledge-work-plugins · 123 tokens

html-ppt-hermes-cyber-terminal

OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.

nexu-io/open-design · 53 tokens

bigquery-ai-ml

Leverages BigQuery's built-in machine learning and GenAI capabilities for advanced data analytics. Use when you need to write SQL queries that perform time-series forecasting, predict values, detect outliers or anomalies, find key drivers, perform semantic search or vector search, classify text, calculate similarity…

google/skills · 104 tokens

mixed-precision

Use FP16/BF16 mixed precision to accelerate training and reduce memory. Use when optimizing GPU performance.

aiming-lab/AutoResearchClaw · 25 tokens

model-compatibility

Model family compatibility matrix covering loaders, resolutions, samplers, CFG, VAE, ControlNet, and LoRA compatibility for SD 1.5, SDXL, Flux, SD3, and video models.

artokun/comfyui-mcp · 47 tokens

civitai

Discover Civitai models with the BUILT-IN downloadmodel action:"searchcivitai" and install/generate them locally. Find a checkpoint/LoRA/embedding on Civitai, download it into ComfyUI, and use its trigger words. Optionally pair the official Civitai MCP for community features (images browsing, posting, collections).

artokun/comfyui-mcp · 76 tokens