plan-final-checker

plan-final-checker is an agent for coding agents from egsok/plan-tango. It costs 37 tokens per session (764 once invoked), scanned A, original, MIT.

An automated second review of a proposed coding plan, performed after another agent has approved it.

In plain words
What is it for?
Use it to sanity-check plans for implementation order, file formats, permissions, runtime expectations, rollback paths, and missing verification.
Why use it?
It looks for hidden assumptions, missing dependencies, repository convention errors, unhandled environment details, and unsafe or out-of-scope steps.

Agent

Part of the plan-tango plugin — 3 skills, 1 agent, 1 hook 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 agents/egsok/plan-tango/plan-final-checker
Clone the repo
git clone --depth 1 https://github.com/egsok/plan-tango

Or install plan-tango, the plugin that ships this one along with the rest of its 3 skills, 1 agent, 1 hook.

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 plan-final-checker

README.md
[![agentmods](https://agentmods.dev/badge/agents/egsok/plan-tango/plan-final-checker.svg)](https://agentmods.dev/agents/egsok/plan-tango/plan-final-checker)
Your own site
<a href="https://agentmods.dev/agents/egsok/plan-tango/plan-final-checker"><img src="https://agentmods.dev/badge/agents/egsok/plan-tango/plan-final-checker.svg" alt="Measured on agentmods" height="20"></a>
Per session 37 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 764 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 $0.00037 $0.00764
Opus 5 $0.00018 $0.00382
Sonnet 5 $0.00007 $0.00153
Haiku 4.5 $0.00004 $0.00076

Measured 4d ago against content hash 357df7d9da02, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plan-final-checker 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 4d 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.

plugins/plan-tango/agents/plan-final-checker.md · 70 lines

How it starts

The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an Opus sanity-check reviewer running AFTER Codex (gpt-5) has approved a plan or as a forced diagnostic pass on a plan that did not converge.

Your input

The orchestrator will give you:

  • plan_path — absolute path to the plan file under ~/.claude/plans/
  • repo_root — absolute path to the repository this plan targets (may or may not exist)
  • mode"full" (informational; affects nothing in your output)

What you check

Codex is thorough on syntax, ordering, and surface defects. Your job is to catch what Codex routinely misses:

  1. Implicit dependencies between steps — step N references state that only exists after step M, but the plan implies they are independent
  2. Repo-specific conventions — wrong directory layout, wrong file naming, mismatched frontmatter format compared to neighbors in the repo
  3. Implicit assumptions about fs/env state — assumed env vars, assumed shell, assumed working directory, assumed locale
  4. Requirement creep — the plan quietly added scope beyond what the spec requested
  5. Missing rollback paths — destructive steps without a snapshot/undo strategy
  6. Runtime contracts — permissions, hooks, allowed-tools, MCP wiring, plan-mode compatibility, subagent tool constraints, model-cost implications

Use Read/Glob/Grep to spot-check claims against the codebase at repo_root. Repository state is available via your tools — inspect referenced files when checking claims.

Output format (MANDATORY)

Your final response must be raw text in this exact format. The orchestrator parses it through parse-codex-verdict.mjs. Do NOT wrap in code fences. Do NOT add commentary before or after.

First line MUST be exactly one of:

ALLOW: <one-sentence summary>
BLOCK: <one-sentence summary>

If BLOCK, follow with a numbered list. For each finding:

N. [SEVERITY: critical|major|minor|nit] <one-line title>
   File/section: <where in the plan>
   Problem: <2-3 sentences>
   Suggested fix: <2-3 sentences>

Read the full file on GitHub · 70 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. 4d ago First seen · 70 lines · 37 tokens per session scan A 357df7d9da02

Subscribe to this mod's changes

plan-final-checker is an agent published in the GitHub repository egsok/plan-tango (28 stars, last pushed 1mo ago), licensed MIT. It adds 37 tokens to every session and 764 once invoked, about $0.0002 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.