Skip to content

Commit 7c6a479

Browse files
authored
docs: document enabling experimental skills (#8024)
## Notes Skills are behind the experimental `skills` feature flag (disabled by default), but the skills guide didn't explain how to turn them on. - Add an explicit enable section to `docs/skills.md` (config + `--enable`) - Add the skills flag to `docs/config.md` and `docs/example-config.md` - Document the `/skills` slash command
1 parent 5d77d4d commit 7c6a479

File tree

4 files changed

+39
-18
lines changed

4 files changed

+39
-18
lines changed

docs/config.md

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -49,6 +49,7 @@ Supported features:
4949
| `ghost_commit` | false | Experimental | Create a ghost commit each turn |
5050
| `enable_experimental_windows_sandbox` | false | Experimental | Use the Windows restricted-token sandbox |
5151
| `tui2` | false | Experimental | Use the experimental TUI v2 (viewport) implementation |
52+
| `skills` | false | Experimental | Enable discovery and injection of skills |
5253

5354
Notes:
5455

docs/example-config.md

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -220,6 +220,7 @@ view_image_tool = true
220220
web_search_request = false
221221
ghost_commit = false
222222
enable_experimental_windows_sandbox = false
223+
skills = false
223224

224225
################################################################################
225226
# Experimental toggles (legacy; prefer [features])

docs/skills.md

Lines changed: 18 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -4,6 +4,19 @@
44
55
Codex can automatically discover reusable "skills" you keep on disk. A skill is a small bundle with a name, a short description (what it does and when to use it), and an optional body of instructions you can open when needed. Codex injects only the name, description, and file path into the runtime context; the body stays on disk.
66

7+
## Enable skills
8+
9+
Skills are behind the experimental `skills` feature flag and are disabled by default.
10+
11+
- Enable in config (preferred): add the following to `$CODEX_HOME/config.toml` (usually `~/.codex/config.toml`) and restart Codex:
12+
13+
```toml
14+
[features]
15+
skills = true
16+
```
17+
18+
- Enable for a single run: launch Codex with `codex --enable skills`
19+
720
## Where skills live
821

922
- Location (v1): `~/.codex/skills/**/SKILL.md` (recursive). Hidden entries and symlinks are skipped. Only files named exactly `SKILL.md` count.
@@ -23,6 +36,11 @@ Codex can automatically discover reusable "skills" you keep on disk. A skill is
2336
- If valid skills exist, Codex appends a runtime-only `## Skills` section after `AGENTS.md`, one bullet per skill: `- <name>: <description> (file: /absolute/path/to/SKILL.md)`.
2437
- If no valid skills exist, the section is omitted. On-disk files are never modified.
2538

39+
## Using skills
40+
41+
- Mention a skill by name in a message using `$<skill-name>`.
42+
- In the TUI, you can also use `/skills` to browse and insert skills.
43+
2644
## Validation and errors
2745

2846
- Invalid skills (missing/invalid YAML, empty/over-length fields) trigger a blocking, dismissible startup modal in the TUI that lists each path and error. Errors are also logged. You can dismiss to continue (invalid skills are ignored) or exit. Fix SKILL.md files and restart to clear the modal.

docs/slash_commands.md

Lines changed: 19 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -10,23 +10,24 @@ Slash commands are special commands you can type that start with `/`.
1010

1111
Control Codex’s behavior during an interactive session with slash commands.
1212

13-
| Command | Purpose |
14-
| ------------ | ----------------------------------------------------------- |
15-
| `/model` | choose what model and reasoning effort to use |
16-
| `/approvals` | choose what Codex can do without approval |
17-
| `/review` | review my current changes and find issues |
18-
| `/new` | start a new chat during a conversation |
19-
| `/resume` | resume an old chat |
20-
| `/init` | create an AGENTS.md file with instructions for Codex |
21-
| `/compact` | summarize conversation to prevent hitting the context limit |
22-
| `/undo` | ask Codex to undo a turn |
23-
| `/diff` | show git diff (including untracked files) |
24-
| `/mention` | mention a file |
25-
| `/status` | show current session configuration and token usage |
26-
| `/mcp` | list configured MCP tools |
27-
| `/logout` | log out of Codex |
28-
| `/quit` | exit Codex |
29-
| `/exit` | exit Codex |
30-
| `/feedback` | send logs to maintainers |
13+
| Command | Purpose |
14+
| ------------ | -------------------------------------------------------------------------- |
15+
| `/model` | choose what model and reasoning effort to use |
16+
| `/approvals` | choose what Codex can do without approval |
17+
| `/review` | review my current changes and find issues |
18+
| `/new` | start a new chat during a conversation |
19+
| `/resume` | resume an old chat |
20+
| `/init` | create an AGENTS.md file with instructions for Codex |
21+
| `/compact` | summarize conversation to prevent hitting the context limit |
22+
| `/undo` | ask Codex to undo a turn |
23+
| `/diff` | show git diff (including untracked files) |
24+
| `/mention` | mention a file |
25+
| `/status` | show current session configuration and token usage |
26+
| `/mcp` | list configured MCP tools |
27+
| `/skills` | browse and insert skills (experimental; see [docs/skills.md](./skills.md)) |
28+
| `/logout` | log out of Codex |
29+
| `/quit` | exit Codex |
30+
| `/exit` | exit Codex |
31+
| `/feedback` | send logs to maintainers |
3132

3233
---

0 commit comments

Comments
 (0)