# Execute

> Execute the code in a Stencila document and save the outputs into it.

- Skill: `stencila/execute` (Agent Skill)
- Install (CLI): `npx skillmds@latest add stencila/execute`
- Raw SKILL.md: https://api.skillmd.com/api/skills/stencila/execute/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: stencila (https://skillmd.com/u/stencila)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/stencila/execute

---


# /stencila:execute

Execute the document given in `$ARGUMENTS`.

Use the stencila CLI path reported in session context; otherwise
`${CLAUDE_PLUGIN_OPTION_STENCILA_PATH}` if set, else `stencila`. If no CLI
is available, point the user at `/stencila:doctor` and stop.

```sh
NO_COLOR=1 <stencila> execute <input> --yes
```

- Execution is staleness-driven: unchanged chunks do not re-run. Add
  `--force-all` when the user wants everything re-executed.
- When the user only wants to *check* the document (not update it), add
  `--no-save` so outputs are not written back.
- `--ignore-errors` continues past failing chunks.

Afterwards, summarise: which chunks ran, which errored (quote the first
error and locate its chunk), and whether the document was saved. See the
`execution` skill for staleness and kernel details.