Plan

Skill Lemonaderrrrr/flow-kit/skills/plan

Stop Claude from winging it — 7 workflow skills that route every task through two thinking modes (Philosopher = absorb / Engineer = ship). Works in Claude Code + Codex.

Install

npx -y skills add Lemonaderrrrr/flow-kit --skill plan

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Turn a goal into an executable plan with milestones, dependencies, schedule, and checkpoints. Default Engineer mode. Uses whatever context the user provides as a baseline, fine-tunes from current info, and asks for missing key inputs instead of assuming. Saves to outputs/plans/. Use when the user says "help me plan X", "make a timeline", "schedule this", or invokes the skill.

SKILL.md

2.2 KB, as published. Nobody here has run it

Plan Workflow

Turn a goal into a milestone-based, dependency-aware, scheduled, trackable plan.

Start

Route-confirm (see CONVENTIONS.md): one line stating this is the plan workflow and the suggested mode; wait for the go-ahead.
Confirm mode: default ⚙️ Engineer; the user may switch to 🧠 Philosopher.
Establish the baseline: use any context the user has provided (notes, prior plans, constraints) as the starting point; fine-tune from current info. When a key input is missing, ask the user before scheduling — don't assume.

⚙️ Engineer mode (default)

Goal: a plan you can execute against.

Define the goal — goal + measurable success criterion + deadline.
Take stock — assemble the starting point from context + known constraints/resources; state your assumptions out loud ("I'm assuming X — correct?") and ask for any missing key facts.
Decompose — milestones → tasks → mark dependencies.
Schedule — timeline, critical path, buffers.
Risks + Plan B — what's most likely to slip, and the fallback.
Checkpoints — how to track progress and when to review.
Capture — Save to outputs/plans/plan_<english-slug>.md.

🧠 Philosopher mode

For thinking before scheduling: should this goal be pursued, what's the real objective? Then switch to Engineer mode to schedule.

Wrap-up conventions

In the plan, separate: known facts / your assumptions / awaiting-confirmation.
Live-search principle: verify external facts (deadlines, policies, requirements) online — don't rely on training memory.
Keep a "current progress + next step" line so work resumes across sessions.
Promote: if it's directional/reusable, offer to save it to long-term memory; one-off outputs stay put.
Filenames use English slugs.