A command that researches and plans a feature or bug fix within defined agent responsibilities. It creates a wave-based plan on a separate work branch for later implementation.

Use it to plan changes such as loading states, timezone fixes, or CSV export. It researches the codebase, writes the plan, creates the work branch, and commits the plan there.

It keeps research summaries out of the main working context and prevents planning changes from being mixed with the default branch. It also makes role limits and outside dependencies explicit.

How seanrreid/RAD_framework configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to seanrreid/RAD_framework. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/seanrreid/RAD_framework/main/.claude/commands/team/rad-plan.md

Clone the repo

git clone --depth 1 https://github.com/seanrreid/RAD_framework

Made for: Claude Code.

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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/seanrreid/rad_framework/rad-plan/github.svg)](https://agentmods.dev/commands/seanrreid/rad_framework/rad-plan)

Your own site

<a href="https://agentmods.dev/commands/seanrreid/rad_framework/rad-plan"><img src="https://agentmods.dev/badge/commands/seanrreid/rad_framework/rad-plan/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/commands/seanrreid/rad_framework/rad-plan"><img src="https://agentmods.dev/badge/commands/seanrreid/rad_framework/rad-plan.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 80 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 5,207 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00080	$0.05207
Opus 5.5	$0.00032	$0.02083
Sonnet 5.5	$0.00016	$0.01041
Haiku 4.5	$0.00008	$0.00521

Measured 17d ago against content hash c204499b5347, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

rad-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 17d 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.

.claude/commands/team/rad-plan.md · 515 lines

How it starts

The opening of the file, as written. The whole thing — 515 lines — stays where its author put it; the contents beside it link to each section on GitHub.

rad-plan

Research and plan a feature within your agent boundaries. The plan doc lives on its own rad/[feature] work branch (Lane B): nothing is committed to the default branch here. The architect approves with rad-approve (/architect:rad-approve in Claude Code) — there is no plan PR. The plan and its code reach the default branch later, together, via the single deliver PR.

Input

The input ($ARGUMENTS) should describe the feature or bug fix. Examples:

"Add skeleton loading states to the habit list"
"Fix the date picker not respecting user timezone"
"Add export to CSV for the weekly summary"

If $ARGUMENTS is empty, ask for a description before proceeding.

--light may prefix the description to request a light-tier plan — a condensed, single-wave plan for a small, low-risk change:

"--light Fix the typo in the onboarding guide"

Without --light, every step below runs in standard mode, unchanged.

--override-disposition "<reason>" may accompany the description to proceed past a research artifact's cited non-actionable disposition (see Step 0). The override and its reason are recorded in the plan header.

--playbook <kind>/<slug>[@<version>] may accompany the description to start from a pinned playbook in .agents/playbooks/ (see Step 1b). A bare ref pins the playbook's current version.

Process

Step 0: Check the research disposition

Derive the feature slug the same way Step 3 does (kebab-case; if $ARGUMENTS names a research slug, use it). If .agents/research/[feature-slug].md does not exist, skip this step — no research artifact, no check.

Otherwise run:

scripts/check-disposition.sh .agents/research/[feature-slug].md
# when $ARGUMENTS includes --override-disposition "<reason>":
scripts/check-disposition.sh .agents/research/[feature-slug].md --override "<reason>"

Exit 1 — cited non-actionable disposition. Stop and show its output (the disposition, the evidence, and the --override-disposition "<reason>" syntax). Do not research and do not write a plan. In an unattended run this is a stop — report and end, never wait for input.
Exit 2 — usage error or unreadable artifact. Report the error and stop.
Exit 0 — proceed. If the output begins overridden:, add Disposition-Override: [disposition] — [reason] to the plan's header block (after Branch:, and after Tier: in a light plan).

Read the full file on GitHub · 515 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.

17d ago Changed · +3 lines c204499b5347
20d ago Changed · +58 lines e4c956b45280
22d ago Changed · -1 lines · +4 tokens per session a11e8832edeb
27d ago Changed · +6 lines 935e2cafa3d4
29d ago Changed · +27 lines b8817a942d93
1mo ago Changed · +12 lines f335d3ab8da4
1mo ago Changed · +130 lines 180a1ba046c3
1mo ago Changed · +4 lines b41d83b861e7
2mo ago First seen · 276 lines · 76 tokens per session scan A 8f2d9dc023dd

Subscribe to this mod's changes

rad-plan is a command published in the GitHub repository seanrreid/RAD_framework (5 stars, last pushed 4d ago), licensed MIT. It adds 80 tokens to every session and 5,207 once invoked, about $0.0003 per session on Opus 5.5. 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.

Planning

RAD_framework: Command for Claude Code

Borrowing it

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

rad-plan

Input

Process

Step 0: Check the research disposition

What this file has done since we first saw it

Other commands, from other repositories

checklist

clarify

specify

converge

implement

speckit.bug.fix