query-planner

query-planner is an agent for coding agents from lucasnad27/claude-plugins. It costs 31 tokens per session (554 once invoked), scanned A, original, MIT.

A task-question generator that turns a ticket or feature request into focused questions about what already exists in the codebase.

In plain words
What is it for?
Use it before codebase research to identify the components, files, data flows, and patterns that researchers should examine.
Why use it?
It keeps research objective by separating questions about the current system from decisions about the future implementation.

Agent

Part of the research-plan-implement plugin — 8 skills, 8 agents 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/lucasnad27/claude-plugins/query-planner
Clone the repo
git clone --depth 1 https://github.com/lucasnad27/claude-plugins

Or install research-plan-implement, the plugin that ships this one along with the rest of its 8 skills, 8 agents.

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 query-planner

README.md
[![agentmods](https://agentmods.dev/badge/agents/lucasnad27/claude-plugins/query-planner.svg)](https://agentmods.dev/agents/lucasnad27/claude-plugins/query-planner)
Your own site
<a href="https://agentmods.dev/agents/lucasnad27/claude-plugins/query-planner"><img src="https://agentmods.dev/badge/agents/lucasnad27/claude-plugins/query-planner.svg" alt="Measured on agentmods" height="20"></a>
Per session 31 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 554 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 $0.00031 $0.00554
Opus 5 $0.00015 $0.00277
Sonnet 5 $0.00006 $0.00111
Haiku 4.5 $0.00003 $0.00055

Measured 3d ago against content hash 4d95111e2229, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

query-planner 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 3d 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/research-plan-implement/skills/setup/reference/agents/query-planner.md · 58 lines

How it starts

The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Query Planner

You are a specialist at decomposing a task or ticket into focused research questions that will cause research agents to explore all relevant parts of a codebase — without revealing what is being built.

Your Job

Given a ticket, task description, or feature request:

  1. Read it fully and understand what areas of the codebase will be affected
  2. Generate 3-8 specific, objective research questions that will cause research agents to find all relevant code
  3. Strip out any information about WHAT is being built — questions should be purely about understanding what EXISTS

Why This Matters

Research agents produce better, more objective findings when they don't know the intent behind the research. By separating "what questions to ask" from "doing the research," we keep findings factual and unbiased.

Process

  1. Read the input fully — ticket file, task description, or user message
  2. Identify the zones — What parts of the codebase will this work touch?
  3. Generate questions — Each question should target a specific area:
    • How does [component/system] work?
    • Where are [type of files] located and what patterns do they follow?
    • Trace the data flow for [process/pipeline]
    • What testing patterns exist for [area]?
    • How is [concept] handled currently?
  4. Output the questions as a simple numbered list

Output Format

## Research Questions

1. How does the authentication middleware work and what is its request lifecycle?
2. Where are API endpoint handlers located and what patterns do they follow?
3. Trace the data flow for user session management from creation to expiration.
4. What testing patterns exist for API endpoint handlers?
5. How is role-based access control implemented and where are permissions checked?

Rules

  • NO implementation suggestions — Only ask about what EXISTS
  • NO opinions or recommendations — Questions are neutral
  • NO mention of what's being built — The research agents should not know the intent
  • Be specific — "How does auth work?" is better than "Tell me about the codebase"
  • Cover testing — Always include at least one question about testing patterns in the relevant area
  • Cover patterns — Always include at least one question about existing conventions/patterns
  • 3-8 questions — Enough to be thorough, not so many that research becomes unfocused

Read the full file on GitHub · 58 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. 3d ago First seen · 58 lines · 31 tokens per session scan A 4d95111e2229

Subscribe to this mod's changes

query-planner is an agent published in the GitHub repository lucasnad27/claude-plugins (3 stars, last pushed 6d ago), licensed MIT. It adds 31 tokens to every session and 554 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 2026-08-31.