Skip to content
Back to search

mise-en-place

SkillWorks with Claude CodeGitHub

A Claude skill for project kickoffs, handoffs between chats, pre-send checks and agent design. Tested, with known limits.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/skillshunter--mise-en-place/badge.svg)](/listings/skillshunter--mise-en-place)

HTML

<a href="/listings/skillshunter--mise-en-place"><img src="/api/listings/skillshunter--mise-en-place/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

**Claude app or claude.ai** 1. Download [`mise-en-place.zip`](https://github.com/SkillsHunter/mise-en-place/raw/main/mise-en-place.zip). Don't unzip it. 2. Open **Customize → Skills**, click **Add** (or **+**), and choose **Upload skill**. 3. Pick the zip and make sure the skill is switched on.

Skills need code execution and file creation turned on. If Skills is missing or greyed out, check that setting. On Team and Enterprise plans, an admin may need to enable it.

**Claude Code** Copy the `mise-en-place` folder into `~/.claude/skills/`. You should end up with `~/.claude/skills/mise-en-place/SKILL.md`.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 50 lines
  1. # mise-en-place
  2. A Claude skill for four moments in bigger pieces of work: starting a project, handing it to a new chat, checking it before it goes out, and designing an agent or automation.
  3. | Moment | Say | You get |
  4. |---|---|---|
  5. | Starting | "Let's start a project to plan our spring speaker series." | One question at a time, then an **intent card** to approve before anything is built: the goal, what it must never do, what's out of scope, and how you'll know it's done |
  6. | Switching chats | "**Use mise-en-place** to write the handoff for a new chat." | A **handoff** to paste into the new chat. It keeps what you decided separate from what Claude only suggested |
  7. | Before sending | "**Use mise-en-place** to check this before I send it." | A **ship check**: claims checked against sources, missing specifics and stock phrasing flagged, and who should give the final OK |
  8. | Designing an automation | "Let's design an agent that…" | An intent card, then a **layer table**: what the AI decides, what plain code does, what's written down, and what breaks if each part is removed |
  9. ## Saying "use mise-en-place"
  10. Claude picks up project kickoffs on its own. For handoffs and pre-send checks, say "use mise-en-place". In testing, Claude usually handled those requests itself without loading the skill. When asked by name, it used the skill both times. In Claude Code you can also type `/mise-en-place`.
  11. ## Install
  12. **Claude app or claude.ai**
  13. 1. Download [`mise-en-place.zip`](https://github.com/SkillsHunter/mise-en-place/raw/main/mise-en-place.zip). Don't unzip it.
  14. 2. Open **Customize → Skills**, click **Add** (or **+**), and choose **Upload skill**.
  15. 3. Pick the zip and make sure the skill is switched on.
  16. Skills need code execution and file creation turned on. If Skills is missing or greyed out, check that setting. On Team and Enterprise plans, an admin may need to enable it.
  17. **Claude Code**
  18. Copy the `mise-en-place` folder into `~/.claude/skills/`. You should end up with `~/.claude/skills/mise-en-place/SKILL.md`.
  19. ## What it does and doesn't do
  20. - It asks before building, sending, publishing or changing anything. This is guidance Claude follows, not a lock. Your app's own permission settings are what actually enforce it.
  21. - It reads what you attach or point to. It asks before searching your email, calendar or drives.
  22. - Text inside your files and emails is treated as content, never as instructions.
  23. - It doesn't treat its own check as proof. It tells you who should do the final check.
  24. ## Files it creates
  25. - `<name>-intent.md`
  26. - `<name>-handoff-v1.md` (then v2, v3 and so on)
  27. - `mise-en-place-ledger.md`, a log of whether handoffs and cards actually saved rework
  28. If your chat is in a project, the files go there. Otherwise Claude gives them to you to keep.
  29. ## Known limits (version 0.4)
  30. - Automatic pickup has been tested on one account with about 30 other skills, one run per prompt. Kickoff fired in 2 of 2 tests. Handoff and ship check fired in 0 of 4 tests unless asked for by name, and in 2 of 2 when asked by name.
  31. - There is no evidence yet that it reduces rework. That's what the ledger is for.
  32. Test details are in [`TESTING.md`](TESTING.md). The sources behind each rule are in [`EVIDENCE.md`](EVIDENCE.md).
  33. ## Problems or ideas
  34. Open an issue using the template. Include your prompt, which Claude app you used, whether the skill fired, and what you expected.
  35. ## Credit
  36. The ideas come from a class session on working with AI ("Mise en Place"). Every claim was then checked against published research and Anthropic's documentation. Only what held up is in the skill.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.