A read-only review process for checking AI-written code, plans, and documents before they are presented as complete.

Use it to validate code, documentation, or plans before shipping, merging, or relying on them.

It detects placeholders, invented APIs, false claims of completion, and other problems that can make AI output unsafe to trust.

How VerificateAI/verificate-mcp-quickstart configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to VerificateAI/verificate-mcp-quickstart. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/VerificateAI/verificate-mcp-quickstart/master/.claude/skills/verificate/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/VerificateAI/verificate-mcp-quickstart

Made for: Claude Code.

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/verificateai/verificate-mcp-quickstart/verificate/github.svg)](https://agentmods.dev/skills/verificateai/verificate-mcp-quickstart/verificate)

Your own site

<a href="https://agentmods.dev/skills/verificateai/verificate-mcp-quickstart/verificate"><img src="https://agentmods.dev/badge/skills/verificateai/verificate-mcp-quickstart/verificate/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/verificateai/verificate-mcp-quickstart/verificate"><img src="https://agentmods.dev/badge/skills/verificateai/verificate-mcp-quickstart/verificate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 92 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 656 The whole file, excluding the scripts and references it only reads on demand.

Security scan B 2 findings. A grade says what 26 rules found in the file — not that it is safe.

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 5.1	$0.00092	$0.00656
Opus 5.5	$0.00037	$0.00262
Sonnet 5.5	$0.00018	$0.00131
Haiku 4.5	$0.00009	$0.00066

Measured 2mo ago against content hash 8cb2be5edaec, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

verificate scanned grade B with 2 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 2mo 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.

Sends data to an external URLmediumData exfiltration

A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.

curl -s https://mcp.verificate.ai/mcp -H "content-type: application/json" -d '{

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s https://mcp.verificate.ai/mcp -H "content-type: application/json" -d '{

.claude/skills/verificate/SKILL.md · 48 lines

How it starts

The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Verificate — verify AI-written work before you trust it

The merge gate for AI-written code, plans, and documents. Deterministic reality gates run first and can veto (no model output can override them); a frontier-model review then scores what survives. Read-only: code is analyzed, never executed.

When to use this skill

Before presenting a substantive code change as finished.
When the user asks to validate / review / verify / gate AI-generated code, a plan, or a document.
When a coding agent is about to accept a diff or write code to disk.

How to use it

Preferred — the MCP server (if connected): call validate_ai_output with the code and validation_type: "code_generation" (use "documentation" for prose, validate_plan for designs). Add the server once with:

claude mcp add --transport http verificate https://mcp.verificate.ai/mcp

Fallback — direct HTTP (no MCP client needed): POST a JSON-RPC tools/call to https://mcp.verificate.ai/mcp:

curl -s https://mcp.verificate.ai/mcp -H "content-type: application/json" -d '{
  "jsonrpc":"2.0","id":1,"method":"tools/call",
  "params":{"name":"validate_ai_output","arguments":{
    "ai_output":"<CODE HERE>","validation_type":"code_generation"}}}'

The response contains result.structuredContent with valid (approve/reject), score, and issues (the findings).

What to do with the verdict

If valid is false (REJECTED), fix every finding and validate again. Never present rejected work or summarize the findings away — a deterministic-gate veto cannot be argued past; the flagged content itself must change.
Tell the user, in one plain sentence, what the gate caught. A caught bug is exactly the outcome they want.

Free tier & privacy

25 validations per machine, no signup, no card, no token. After that, add Authorization: Bearer <token> (30-day trial at https://verificate.ai/auth/signup). Code is analyzed, never executed, never used to train models. Privacy: https://verificate.ai/privacy

Read the full file on GitHub · 48 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.

2mo ago First seen · 48 lines · 92 tokens per session scan B 8cb2be5edaec

Subscribe to this mod's changes

verificate is a skill published in the GitHub repository VerificateAI/verificate-mcp-quickstart (1 stars, last pushed 1mo ago), licensed MIT. It adds 92 tokens to every session and 656 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it B with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Code review ai ai-code-review claude code-quality cursor mcp

norma-workflow

Run every coding task through Norma's deterministic checks and record the outcome to the audit trail. Use whenever you create or modify code in a workspace connected to the Norma MCP server, when asked to review code against organization standards, or when asked to work through the repository's open issues.

qualityclouds/norma-mcp · 61 tokens

new-track

Start a new feature or bug track. Collaborative intake process with structured questions, AI guidance, and progressive refinement before generating spec.md and plan.md. Use when the user asks to 'start a new track', 'create a feature track', 'add a bug fix track', or says 'I want to build X', 'fix the Y bug', 'plan a…

drafthq/draft · 78 tokens

verificate-mcp-quickstart: Skill for Claude Code