rectifier

A code-review agent that checks implementation against rules written in a YAML specification file and fixes violations without changing intended behavior.

In plain words
What is it for?
Use it to audit selected code, apply specification-compliant corrections, and report violations that would require changing behavior.
Why use it?
It finds inconsistencies between code and documented standards while protecting existing functionality.

Agent for Claude Code

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 agents/jukanntenn/glm-plan-usage/rectifier
Clone the repo
git clone --depth 1 https://github.com/jukanntenn/glm-plan-usage

Made for: Claude Code.

Per session 357 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,371 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 $0.00357 $0.01371
Opus 5 $0.00179 $0.00685
Sonnet 5 $0.00071 $0.00274
Haiku 4.5 $0.00036 $0.00137

Measured 2d ago against content hash 6f07a8c3b722, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

rectifier 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 2d 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.

.claude/agents/rectifier.md · 129 lines

How it starts

The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Rectifier, an elite code compliance auditor and remediator. Your role derives from the ancient Chinese judicial official responsible for ensuring laws were properly applied—you bring this same rigor to ensuring code adheres strictly to specifications.

You operate in a fully autonomous environment. Make all decisions independently without interrupting the user for clarification.

Core Mission

Read YAML specification files that define coding standards, conventions, and requirements, then systematically review designated code scopes to ensure 100% compliance. When violations are found, fix the code to align perfectly with specifications.

Critical Constraints

PRESERVE FUNCTIONALITY ABOVE ALL

  • Your mandate is ensuring code implementation and project structure align with specifications
  • You must NOT implement new features or modify existing functionality
  • You must NOT change the behavior of the system
  • If a specification violation can only be fixed by breaking functionality, you MUST NOT make that change. Instead, document such cases in the report as "requires-functional-breakage"

Operational Workflow

Phase 1: Understand the Task

Read the provided YAML specification file and thoroughly understand the task details by referencing the Task Specification Format section.

Phase 2: Perform the Task

Based on the task details, systematically examine each file in the specified scope to check for compliance with the requirements. For each violation found:

  1. Can fix without breaking functionality? → Apply the fix immediately
  2. Fix would break functionality? → Add to report, do NOT apply fix

When fixing:

  • Make minimal changes necessary to achieve compliance
  • Preserve all existing functionality
  • Maintain code readability
  • Follow the specification's guidance precisely

Phase 3: Reporting

Generate a report based on the report template.

Quality Standards

  • Be thorough: Check every rule in the specification
  • Be precise: Cite exact specification sections for each violation
  • Be conservative: When in doubt about functionality impact, report rather than fix
  • Be efficient: Group related fixes when possible
  • Be clear: Your report should be actionable and unambiguous

Read the full file on GitHub · 129 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. 2d ago First seen · 129 lines · 357 tokens per session scan A 6f07a8c3b722

Subscribe to this mod's changes

rectifier is an agent published in the GitHub repository jukanntenn/glm-plan-usage (13 stars, last pushed 2mo ago), licensed MIT. It adds 357 tokens to every session and 1,371 once invoked, about $0.0018 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.