Plan Mode

What it does

When your agent uses it

• The user says /plan
• Think before coding
• Asks the agent to plan before executing
• Infrastructure changes

Example prompts

• “plan mode”
• “plan first”
• “think before coding”
• “/plan-mode”

Workflow steps

6 steps, taken from the step headings in SKILL.md.

1. 1Understanding (What)
2. 2Approach Options (How)
3. 3Task Breakdown (Steps)
4. 4Risk Assessment
5. 5Files & Scope
6. 6Approval Gate

What it can do on your machine

Read from SKILL.md and the folder at commit 79d019b. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are yaml).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Plan Mode loads about 2.1k tokens when it runs. Until then it costs about 185 tokens; SKILL.md has 787 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~185

When it runs · the whole SKILL.md, loaded when a task matches

~2.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from espennilsen/pi at commit 79d019b, republished under its MIT licence (© espennilsen). 787 words, ~2,086 tokens.

name

plan-mode

description

Structured planning mode for AI agents working on coding and development tasks. Produces a comprehensive plan BEFORE any code is written or changed, including task decomposition, architecture decisions, risk assessment, and approval gates. Use this skill whenever the user says "/plan", "plan mode", "plan first", "think before coding", "make a plan", or asks the agent to plan before executing. Also trigger when the user describes a complex multi-step development task and would clearly benefit from structured planning before implementation — especially for tasks involving new features, refactors, migrations, infrastructure changes, or multi-file edits. When in doubt, plan. A few minutes of planning prevents hours of rework.

## Understanding
[Restate the goal in 2-3 sentences]

**Assumptions:**
- [List anything you're assuming that wasn't explicitly stated]

**Open questions:**
- [List anything you need clarified before proceeding]

## Approach Options

### Option A: [Name]
**Summary:** [One line]
**How it works:** [3-5 bullet points, high-level]
**Pros:** [What's good about this]
**Cons:** [Tradeoffs]
**Effort:** [small / medium / large]

### Option B: [Name]
...

**Recommendation:** Option [X] because [reason tied to project context].

## Task Breakdown

### Phase 1: [Name, e.g. "Setup / Preparation"]
- [ ] Step 1.1: [Concrete action]
- [ ] Step 1.2: [Concrete action]
  **Checkpoint:** [What should be true after this phase — a testable condition]

### Phase 2: [Name, e.g. "Core Implementation"]
- [ ] Step 2.1: [Concrete action]
- [ ] Step 2.2: [Concrete action]
  **Checkpoint:** [Testable condition]

### Phase 3: [Name, e.g. "Verification & Cleanup"]
- [ ] Step 3.1: [Concrete action]
- [ ] Step 3.2: [Concrete action]
  **Checkpoint:** [Testable condition]

## Risks

| Risk | Likelihood | Impact | Mitigation |
|------|-----------|--------|------------|
| [Specific thing that could go wrong] | low/med/high | low/med/high | [How to prevent or handle it] |

## Affected Files
- **Create:** `path/to/new-file.ts` — [purpose]
- **Modify:** `path/to/existing-file.ts` — [what changes]
- **Delete:** `path/to/old-file.ts` — [why]

## Ready to Execute?

Plan is complete. Please review and let me know:
- ✅ **Approve** — proceed with implementation
- 🔄 **Revise** — adjust the approach (tell me what to change)
- ❓ **Clarify** — answer open questions first
- ❌ **Cancel** — abort this task

---
task: "[One-line task description]"
status: proposed  # proposed → approved → in-progress → completed → cancelled
created: YYYY-MM-DD
approach: "[Name of chosen/recommended approach]"
---

© espennilsen, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/plan-mode of espennilsen/pi.

Open the folder on GitHubat commit 79d019b