stage-7-verification

stage-7-verification is a skill for Claude Code from cdeust/ai-architect-mcp. It costs 5 tokens per session (2,770 once invoked), scanned A, original, MIT.

A deterministic verification stage that checks code and visual output against fixed rules, structural checks, and build requirements. It reports only pass or fail and does not use language-model judgment or fix problems.

In plain words
What is it for?
Use it to verify an implementation after coding, including visual regression checks, structural validation, and build gates. It requires an implementation record from the previous stage.
Why use it?
It removes uncertainty about whether an implementation meets its required checks. It also makes visual regressions, failed builds, and rule violations explicit.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the ai-architect plugin — 13 skills, 4 commands, 2 MCP servers, 1 plugin shipped together

Good fit Use it to verify an implementation after coding, including visual regression checks, structural validation, and build gates. It requires an implementation record from the previous stage.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/cdeust/ai-architect-mcp/stage-7-verification
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 cdeust/ai-architect-mcp --skill stage-7-verification
Clone the repo
git clone --depth 1 https://github.com/cdeust/ai-architect-mcp

Made for: Claude Code.

Or install ai-architect, the plugin that ships this one along with the rest of its 13 skills, 4 commands, 2 MCP servers, 1 plugin.

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 stage-7-verification

README.md
[![agentmods](https://agentmods.dev/badge/skills/cdeust/ai-architect-mcp/stage-7-verification/github.svg)](https://agentmods.dev/skills/cdeust/ai-architect-mcp/stage-7-verification)
Your own site
<a href="https://agentmods.dev/skills/cdeust/ai-architect-mcp/stage-7-verification"><img src="https://agentmods.dev/badge/skills/cdeust/ai-architect-mcp/stage-7-verification/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 stage-7-verification

Your own site · 80×15
<a href="https://agentmods.dev/skills/cdeust/ai-architect-mcp/stage-7-verification"><img src="https://agentmods.dev/badge/skills/cdeust/ai-architect-mcp/stage-7-verification.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 5 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,770 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.
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 $0.00005 $0.02770
Opus 5 $0.00003 $0.01385
Sonnet 5 $0.00001 $0.00554
Haiku 4.5 $0.00001 $0.00277

Measured 10d ago against content hash a6370a1485f6, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

stage-7-verification 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/stage-7-verification/SKILL.md · 283 lines

How it starts

The opening of the file, as written. The whole thing — 283 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Allostatic Priming

You are a deterministic validator. No judgment. No interpretation. Rules run. Vision diffs. Build compiles or it does not. You do not fix — you report. Every check is binary: pass or fail. You run 64 HOR rules, structural checks, and build gates. No LLM calls. No fuzzy matching. No "close enough."

Trigger

USE WHEN: verify code, run gates, check implementation, HOR rules, visual regression, 64 rules, deterministic, no LLM, structural check, build gate, binary validation, compliance check NOT FOR: code implementation — stage 6, benchmark — stage 8, any LLM reasoning

Survival Question

"Do all 64 HOR rules pass, does the build succeed, and are there no visual regressions?"

Before you start

  1. ai_architect_load_context(stage_id=6, finding_id="{findingID}") — load implementation manifest from Stage 6
  2. ai_architect_load_session_state(session_id="{sessionID}") — confirm currentStage = 7, check retryCount

Missing Stage 6 implementation manifest = BLOCK. Cannot verify without implementation.

CRITICAL: This stage is fully deterministic. No LLM calls. No ai_architect_enhance_prompt. No ai_architect_select_strategy. No ai_architect_expand_thought. Only deterministic tools from the allowlist below.

Allowed tools (closed allowlist)

Only these tools may be called in Stage 7:

  • ai_architect_run_hor_rules
  • ai_architect_run_hor_category
  • ai_architect_run_hor_single
  • ai_architect_run_build
  • ai_architect_run_tests
  • ai_architect_compound_score
  • ai_architect_verify_graph
  • ai_architect_load_context
  • ai_architect_save_context
  • ai_architect_save_session_state
  • ai_architect_append_audit_event
  • ai_architect_emit_ooda_checkpoint
  • ai_architect_fs_read
  • ai_architect_fs_write
  • ai_architect_fs_list
  • ai_architect_git_diff

Any tool not on this list = BLOCK. Enforced by hooks/pre-tool-use/stage-7-gate.sh.

Input contract

Field Type Source Required
stage-6-implementation-manifest.json JSON StageContext[stage-6] YES — BLOCK if missing
Implementation branch git ref pipeline/{findingID} YES — code must exist
PRD files MD StageContext[stage-4] YES — for HOR rule evaluation

Read the full file on GitHub · 283 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 · 283 lines · 5 tokens per session scan A a6370a1485f6

Subscribe to this mod's changes

stage-7-verification is a skill published in the GitHub repository cdeust/ai-architect-mcp (1 stars, last pushed 4mo ago), licensed MIT. It adds 5 tokens to every session and 2,770 once invoked, about $0.0000 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.