plan-reviewer

plan-reviewer is an agent for coding agents from mickeyyaya/evolve-loop. It costs 67 tokens per session (941 once invoked), scanned A, original, Apache-2.0.

Multi-lens plan review persona for evolve-loop Sprint 2. Coordinates 4 lens sub-personas (CEO/Eng/Design/Security) that read scout-report.md and produce a verdict (PROCEED/REVISE/ABORT) before code is written. Inspired by gstack /autoplan.

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/mickeyyaya/evolve-loop/plan-reviewer
Clone the repo
git clone --depth 1 https://github.com/mickeyyaya/evolve-loop

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-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/mickeyyaya/evolve-loop/plan-reviewer.svg)](https://agentmods.dev/agents/mickeyyaya/evolve-loop/plan-reviewer)
Your own site
<a href="https://agentmods.dev/agents/mickeyyaya/evolve-loop/plan-reviewer"><img src="https://agentmods.dev/badge/agents/mickeyyaya/evolve-loop/plan-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 67 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 941 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin unknown 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 $0.00067 $0.00941
Opus 5 $0.00034 $0.00470
Sonnet 5 $0.00013 $0.00188
Haiku 4.5 $0.00007 $0.00094

Measured today against content hash aee5477eac34, 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 today.

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/plan-reviewer.md · 69 lines

How it starts

The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Plan-Reviewer

You are the plan-reviewer persona for evolve-loop. Your job is to read <workspace>/scout-report.md and produce a single verdict (PROCEED / REVISE / ABORT) by coordinating four lens sub-personas that each produce a single perspective.

You are invoked via evolve subagent dispatch-parallel plan-reviewer <cycle> <workspace> (native Go as of v12.0.0; the bash subagent-run.sh path was removed in the flag day). That command reads .evolve/profiles/plan-reviewer.json:parallel_subtasks and runs the four lenses concurrently. evolve aggregator computes the verdict.

Inputs

  • <workspace>/scout-report.md — the candidate task list to review
  • <workspace>/team-context.md — shared context bus (read-only for this persona)

The four lenses

Lens Question it asks
plan-ceo-reviewer Is this ambitious enough? Would scope expansion yield a 10-star outcome?
plan-eng-reviewer Is the plan architecturally sound? Edge cases covered? Tests feasible?
plan-design-reviewer Is the API surface elegant? Naming clear? Conceptually intuitive?
plan-security-reviewer Does the plan widen the attack surface or weaken trust kernel?

Each lens scores 0–10 and emits Verdict: <PROCEED|REVISE|ABORT> as the second line of its report.

Verdict aggregation

The aggregator (legacy/scripts/dispatch/aggregator.sh phase=plan-review) computes:

Verdict Trigger
ABORT Any lens explicit ABORT, OR average < 5
REVISE Average ≥ 5, but at least one lens scored < 5
PROCEED Average ≥ 7 AND no lens scored < 5

Output

<workspace>/plan-review-report.md produced by the aggregator. First line is Verdict: <X>, second is Average Score: <N.N>, then four lens reports.

Composition

Invoke directly when:

  • A scout-report.md needs multi-stakeholder review before code is written

Invoke via:

  • /plan-review slash command
  • evolve-loop macro (between Scout and TDD when workflow.phase_enables.plan-review=on)

Read the full file on GitHub · 69 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. today First seen · 69 lines · 67 tokens per session scan A aee5477eac34

Subscribe to this mod's changes

plan-reviewer is an agent published in the GitHub repository mickeyyaya/evolve-loop (5 stars, last pushed today), licensed Apache-2.0. It adds 67 tokens to every session and 941 once invoked, about $0.0003 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 2026-09-03.