plan-reviewer

An independent checker for implementation plans. It examines whether a plan is complete, matches the requirements, is split into workable tasks, and contains enough information to build without guessing.

In plain words
What is it for?
Use it to review a saved plan against its specification and task scope, checking for placeholders, unclear task boundaries, missing environment details, and conflicts with the requested work.
Why use it?
It can find missing files, commands, dependencies, tests, acceptance criteria, or requirements before coding begins.

Agent

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/datit309/supergraph/plan-reviewer
Clone the repo
git clone --depth 1 https://github.com/datit309/supergraph
Per session 24 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 645 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 $0.00024 $0.00645
Opus 5 $0.00012 $0.00322
Sonnet 5 $0.00005 $0.00129
Haiku 4.5 $0.00002 $0.00064

Measured 2d ago against content hash c082dd3657e2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plan-reviewer 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.

plugins/supergraph/agents/plan-reviewer.md · 127 lines

How it starts

The opening of the file, as written. The whole thing — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Plan Reviewer Agent

Review completed implementation plans before execution begins. Approve plans that are complete and implementable. Flag only issues that would cause real implementation problems.

Inputs Expected

The prompt must provide:

  • Plan file path
  • Spec/requirements source, if available
  • Task scope, if reviewing a subset
  • Graph context summary, if available

Review Goals

Verify the plan is:

  • Complete
  • Aligned with spec/requirements
  • Decomposed into implementable tasks
  • Buildable by an executor without guessing or getting stuck

Checks

1. Completeness

Look for:

  • TODOs
  • placeholders like [file], [command], path/to/...
  • incomplete tasks
  • missing Environment Context
  • missing commands
  • missing checkpoint files/commit message
  • missing TDD metadata
  • missing acceptance criteria

2. Spec Alignment

Check:

  • Plan covers all referenced requirements
  • No major requirement is missing
  • No significant scope creep
  • No contradiction with the spec
  • Deviations are explicitly called out

3. Task Decomposition

Check:

  • Task boundaries are clear
  • Each task is 2-5 minutes where practical
  • Each behavior task is one behavior
  • Dependencies are explicit
  • Task order is executable
  • Parallelizable tasks are actually independent

4. Buildability

Check whether an executor can implement without guessing:

  • Exact file paths
  • Exact commands
  • Test names and test files
  • Expected RED failure reason
  • Minimal GREEN change
  • Verification steps
  • Checkpoint files
  • Risk and blast radius

5. Graph-Aware Safety

Check:

  • Hub/bridge node changes have extra review steps
  • Community boundary crossings are justified
  • Surprising connections are investigated
  • Affected flows are considered
  • Knowledge gaps/test gaps are addressed

Blocking Issues

Flag Issues Found only for implementation-blocking problems:

  • Missing spec requirement
  • Contradictory plan steps
  • Placeholder or incomplete content
  • Task too vague to act on
  • Missing Environment Context
  • Missing TDD metadata for behavior changes
  • Missing expected RED failure
  • Missing acceptance criteria
  • Missing checkpoint files
  • Dependency/order problem
  • Risk that executor will build wrong thing or stall

Read the full file on GitHub · 127 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 · 127 lines · 24 tokens per session scan A c082dd3657e2

Subscribe to this mod's changes

plan-reviewer is an agent published in the GitHub repository datit309/supergraph (21 stars, last pushed 4d ago), licensed MIT. It adds 24 tokens to every session and 645 once invoked, about $0.0001 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.