q-plan

A planning tool that asks several other coding agents for separate implementation plans while your current agent makes its own plan.

In plain words
What is it for?
Use it before starting a coding task when you want alternative plans and a second opinion.
Why use it?
It reduces the risk of committing to one approach without hearing other viewpoints.

Skill for Claude CodeCodex

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 skills/sdewell/code-quorum/q-plan
Any agent
npx skills add sdewell/code-quorum --skill q-plan
Clone the repo
git clone --depth 1 https://github.com/sdewell/code-quorum

Made for: Claude Code, Codex.

Per session 54 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
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 $0.00054 $0.01371
Opus 5 $0.00027 $0.00685
Sonnet 5 $0.00011 $0.00274
Haiku 4.5 $0.00005 $0.00137

Measured yesterday against content hash ade6219c3359, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

q-plan 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 yesterday.

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/code-quorum/skills/q-plan/SKILL.md · 85 lines

How it starts

The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Q-Plan — parallel plans from each registered agent

Completion gate

For every council start, q_await is the blocking completion notification. After completing the host's independent work, call it in the same turn as the start. Do not end the turn, tell the user you will check later, or leave a live job_id pending. If the council is still running, remain in the blocking call until it returns a result or error.

Arguments

$ARGUMENTS is the task description, optionally with --verbose to lift the terse output caps and/or --gemini-model <id> to run the gemini seat on a specific agy model. Strip --verbose and --gemini-model <id> from the task text before passing task. If the remaining task is empty, ask the user what they want a plan for.

Host routing

Claude Code uses the mcp__plugin_code-quorum_quorum__* tools and passes host: "claude". Codex uses the mcp__quorum_codex__* tools and passes host: "codex". The workflows are otherwise identical.

If either required MCP tool is absent from the current task's tool registry, stop and report that the plugin MCP server did not load. Do not substitute the synchronous CLI, direct Python imports, or another runtime path: only the registered start/await pair enforces the structural anti-bias gate.

Every MCP council start requires an explicit absolute cwd. Never omit it: the MCP server's own working directory is its installation or runtime location, not the project selected by the user.

Step 1 — Start the agents (non-blocking)

Call mcp__plugin_code-quorum_quorum__q_plan_start (Claude Code) or mcp__quorum_codex__q_plan_start (Codex) with:

  • task: the value of $ARGUMENTS
  • cwd: the absolute path of the project working directory
  • host: the current host selected above
  • verbose: false by default — plans come back terse (no padding, path:line over pasted code). Set true only if the user gave --verbose.
  • gemini_model: omit by default. Pass an agy model id (exactly as printed by agy models) when the user gave --gemini-model <id> or asked for a specific model in the gemini seat — e.g. claude-opus-4-6-thinking to have that seat answer with Claude on the same AI Pro plan (useful when Gemini quota is tight or a Claude perspective is wanted).

Read the full file on GitHub · 85 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. yesterday First seen · 85 lines · 54 tokens per session scan A ade6219c3359

Subscribe to this mod's changes

q-plan is a skill published in the GitHub repository sdewell/code-quorum (0 stars, last pushed 2d ago), licensed MIT. It adds 54 tokens to every session and 1,371 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 2026-08-31.