Skip to main content
command CX Codex

plan · trailofbits — Claude Code command: what it does, how to install

Creates task_plan.md, findings.md and progress.md in the project root, so the goal, research and session log sit next to the code.

CC-BY-SA-4.0 ★ 540 updated 2026-07-14

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add trailofbits/skills-curated
/plugin install planning-with-files@skills-curated
Open on GitHub

What plan does

First it checks the project root for task_plan.md, findings.md and progress.md. If any of the three is already there, the command lists what it found and waits for a yes before starting over. Then it asks what the task is for and puts that answer into the Goal section of task_plan.md.

Templates come from {baseDir}/skills/planning-with-files/references/templates.md. The session date lands in progress.md, and phase titles appear only when your description of the goal is detailed enough to split into steps. The run ends with a summary of the created files and the current phase, Phase 1 - Requirements & Discovery.

Three tools are allowed: Read, Write and Glob. Nothing runs, no source file is edited, so the plan stays a text you keep working in. Give a fuller answer about the goal, because the phases you get back depend on it.

What it helps with

  • keep the task goal, research notes and session log in separate files
  • start a new task without silently losing the old plan
  • sketch the phases before writing any code
  • record decisions while the research is still fresh
  • log each session with its date and current phase

Example prompts for the agent

  • /plan The goal is OAuth login for the API, split it into phases.
  • Run /plan for the parser rewrite and keep the research notes in findings.md.
  • /plan for the payments module: the goal is test coverage.

How to install plan

1Plugin marketplace

/plugin marketplace add trailofbits/skills-curated
/plugin install planning-with-files@skills-curated

2Manually: put the files in place

.claude/commands/planning-with-files-plan.md

The built-in /plan overrides a file with the same name, so the file is renamed. Call it with /planning-with-files-plan. For all projects use ~/.claude/commands/

Questions about plan

How do I install plan in Claude Code?

Run /plugin marketplace add trailofbits/skills-curated to add the marketplace, then /plugin install planning-with-files@skills-curated.

What does the plan command do?

It creates task_plan.md, findings.md and progress.md in the project root after asking for the task goal, then prints a summary with the current phase.

Will it overwrite planning files I already have?

Not without asking. It checks the root first and shows which of the three files exist before you confirm a fresh start.

Similar slash commands

All slash commands for Agent setup: memory, context, orchestration →

plan by other authors comes first.

plan OthmanAdi Loads the planning-with-files skill and starts planning in the current project. Handy when you want to begin a new plan or continue an existing one. CX Codex +1 ★ 28,179repo stars plan OthmanAdi Starts Manus-style file-based planning for complex engineering workflows. Generates task_plan.md, findings.md, and progress.md or restores an active session. CX Codex +1 ★ 28,179repo stars plan Affaan Mustafa Command invokes a planner sub-agent that breaks down the task into phases, assesses risks, and waits for confirmation before touching any code. CX Codex +1 ★ 6,739repo stars plan disler Turns a task description into a step-by-step implementation plan saved under specs/. Useful when a feature needs thinking through before any code. CX Codex +1 ★ 3,990repo stars plan CloudAI-X Creates and maintains a structured PLAN.md file in your project root to track phases, task dependencies, and progress logs. CX Codex +1 ★ 1,424repo stars planning davila7 planning keeps markdown files for goals, findings and session progress so an agent doesn't drift off the task mid-run. CX Codex +1 712installs plan-goal OthmanAdi plan-goal builds a completion condition out of task_plan.md and feeds it to /goal, so the agent keeps moving through the whole plan. CX Codex +1 ★ 28,179repo stars alan-plan tryalan-ai alan-plan walks the agent through planning: analyze the task, ask clarifying questions, draft the plan and persist it as an Alan artifact. CX Codex +1 1,800installs plan-harder am-will Turns complex requests into structured sprint plans with atomic tasks. Triggers when you tell the agent to plan harder, focusing purely on architecture and risks. CX Codex +1 1,200installs log davila7 The /orchestration/log command sends task completion reports to Linear, Obsidian, Jira or GitHub Issues. Triggered manually after finishing a task or at day's end for a summary. CX Codex +1 351installs exec-plan pproenca exec-plan manages self-contained Markdown execution plans across agent sessions for tasks touching more than three files or taking over an hour. CX Codex +1 239installs

From the same set51

The whole set →
ghidra-headless ghidra-headless is a Claude Code skill that runs Ghidra analysis on compiled binaries without its GUI. It kicks in for decompilation, string and symbol extraction, or call-graph inspection. CX Codex +1 614installs status status reads task_plan.md to display phase milestones and summarizes notes from findings.md. CX Codex +1 ★ 540repo stars openai-gh-address-comments openai-gh-address-comments — a Claude Code skill that collects and addresses reviewer comments on an open PR for the current branch using gh CLI. It fetches all review threads, summarises each one, and lets you pick which to fix. CX Codex +1 ★ 540repo stars grilling The grilling skill questions a plan or design one decision at a time, surfacing unresolved choices and dependencies. Use it before implementation when the details still need agreement. CX Codex +1 ★ 540repo stars openai-yeet openai-yeet takes your changes from a dirty working tree to a draft pull request in one run through GitHub CLI. Handy when the whole chain has to happen at once. CX Codex +1 ★ 540repo stars openai-sentry The openai-sentry Claude Code skill reads Sentry issues and events through the API and summarizes production errors in a compact report. CX Codex +1 ★ 540repo stars python-code-simplifier python-code-simplifier refines recently modified Python code for clarity and maintainability without altering existing program logic or outputs. CX Codex ★ 540repo stars openai-playwright openai-playwright is a Claude Code skill that controls a real browser from the terminal. It runs for page navigation, form entry, data collection, screenshots, and UI debugging. CX Codex +1 ★ 540repo stars

Project sponsors

Companies that keep this analytics open