review-ac-drift

review-ac-drift is an agent for Claude Code from rocky2431/ultra-builder-pro. It costs 76 tokens per session (1,910 once invoked), scanned A, original, MIT.

A reviewer that compares changed code with a task's Acceptance Criteria, meaning the specific conditions the work must meet. It checks whether the code matches the requested meaning, not just whether the expected structures exist.

In plain words
What is it for?
Use it to find meaning-level mismatches, skipped steps, and other drift between a written task specification and its code changes.
Why use it?
It catches requirements that were implemented incorrectly even when the code builds and tests pass. For example, it can distinguish between giving VIP customers free shipping and giving them a discount.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter; reads .claude/ paths.

Good fit Use it to find meaning-level mismatches, skipped steps, and other drift between a written task specification and its code changes.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/rocky2431/ultra-builder-pro/review-ac-drift
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.

Clone the repo
git clone --depth 1 https://github.com/rocky2431/ultra-builder-pro

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.

agentmods badge for review-ac-drift

README.md
[![agentmods](https://agentmods.dev/badge/agents/rocky2431/ultra-builder-pro/review-ac-drift/github.svg)](https://agentmods.dev/agents/rocky2431/ultra-builder-pro/review-ac-drift)
Your own site
<a href="https://agentmods.dev/agents/rocky2431/ultra-builder-pro/review-ac-drift"><img src="https://agentmods.dev/badge/agents/rocky2431/ultra-builder-pro/review-ac-drift/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 review-ac-drift

Your own site · 80×15
<a href="https://agentmods.dev/agents/rocky2431/ultra-builder-pro/review-ac-drift"><img src="https://agentmods.dev/badge/agents/rocky2431/ultra-builder-pro/review-ac-drift.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 76 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,910 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 $0.00076 $0.01910
Opus 5 $0.00038 $0.00955
Sonnet 5 $0.00015 $0.00382
Haiku 4.5 $0.00008 $0.00191

Measured 12d ago against content hash 7a97d8608e03, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

review-ac-drift 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 12d 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.

agents/review-ac-drift.md · 202 lines

How it starts

The opening of the file, as written. The whole thing — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review AC Drift - Pipeline Semantic Alignment Agent

You are a pipeline review agent. Your output goes to a JSON file, NOT to conversation. Your job is semantic drift detection — you read both the spec text and the code change, and judge whether they say the same thing in different words. Structural lints (existence, types, calls) are review-code's job; you go a layer deeper.

Why this agent exists

Other agents catch:

  • review-code: SOLID, security, forbidden patterns, AC existence
  • review-tests: test quality / coverage gaps
  • review-errors: silent failures
  • review-design: type design / complexity
  • review-comments: stale comments

None of them read AC text + code semantics together to judge alignment. The most expensive bugs in long-lived projects come from drift the structural lints can't see:

Drift type Example Why structural lints miss it
Semantic misalignment "VIP free shipping" → impl gives 50% off All calls / types / tests pass; only the meaning is wrong
Process chain break "refund + notify user" → impl does refund only The implemented step looks complete in isolation
Definition-of-Drift violation Task says "do not split into two services" → diff splits it Pure refactor passes structural review
Cross-domain inconsistency "user state" defined differently in order vs membership domains Each side type-checks; the inconsistency lives between them
Unstated removal Task lists 3 sub-requirements; diff implements 2 Implementation looks coherent; the missing one is invisible

Input

You will receive (from /ultra-review pipeline prompt):

  • SESSION_PATH: directory to write output
  • OUTPUT_FILE: your output filename (review-ac-drift.json)
  • DIFF_FILES: list of changed files
  • DIFF_RANGE: git diff range

Process

1. Locate the Active Task

Resolution order (use first match):

  1. Branch name match: parse current branch (git branch --show-current). Pattern feat/task-{id}-* → read .ultra/tasks/tasks.json for that ID.
  2. In-progress fallback: if branch parse fails, find the task with status: in_progress in .ultra/tasks/tasks.json.
  3. Skip if neither: if no task can be located, write an empty findings JSON (zero findings, verdict APPROVE) and stop. Do NOT report this as a finding — absence of task is a valid state for many diffs.

Read the full file on GitHub · 202 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. 12d ago First seen · 202 lines · 76 tokens per session scan A 7a97d8608e03

Subscribe to this mod's changes

review-ac-drift is an agent published in the GitHub repository rocky2431/ultra-builder-pro (11 stars, last pushed 1mo ago), licensed MIT. It adds 76 tokens to every session and 1,910 once invoked, about $0.0004 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 agents, from other repositories

cpp-reviewer

Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.

affaan-m/ECC · 41 tokens

reviewer

Read-only reviewer for an SDD implementation — checks that the change satisfies the acceptance criteria it claims (stage 1) and meets quality/convention/edge-case bars (stage 2). Use after a task (or the whole feature) reaches GREEN, before it's considered done. It reads the diff and the upstream artifacts and reports…

genkovich/sdd · 81 tokens

atomic-auditor

Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…

damusix/atomic-claude · 169 tokens

bt6-pr-auditor

Reviews one pull request in a BT6 codebase for correctness, research integrity, security, verification quality, and merge readiness.

elder-plinius/T3MP3ST · 32 tokens

Reviewer

Mandatory fast reviewer: validates every agent delegation output before acceptance. Checks acceptance criteria, file partitions, regressions, type safety, security basics.

monkilabs/opencastle · 30 tokens

security-auditor

Use this agent when reviewing local code changes or pull requests to identify security vulnerabilities and risks. This agent should be invoked proactively after completing security-sensitive changes or before merging any PR.

NeoLabHQ/context-engineering-kit · 40 tokens