Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 11 additions & 0 deletions .claude/settings.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
{
"extraKnownMarketplaces": {
"dock": {
"source": { "source": "github", "repo": "docknetwork/dock-claude-skills" },
Comment thread
gasher marked this conversation as resolved.
"autoUpdate": true
}
},
"enabledPlugins": {
"dock-skills@dock": true
}
}
20 changes: 20 additions & 0 deletions .claude/skills/example-skill/SKILL.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,20 @@
---
name: example-skill
description: "TEMPLATE — copy this directory, rename it, and write a real trigger here before use. Left disabled so it never fires as-is."
disable-model-invocation: true
---

# example-skill

Boilerplate for a repo-specific Claude Code skill. Copy this directory to
`.claude/skills/<your-skill-name>/`, then:

1. Rename `name` above to match the directory.
2. Replace `description` with the real trigger: the situations that should make
an agent reach for this skill, phrased as explicit branches ("use when the
user says X" / "use after Y fails"), not a vague "use for X".
3. Remove `disable-model-invocation: true` once the skill should fire on its
own — keep it only for a skill meant to be invoked by name alone.
4. Replace this body with real steps, gotchas, and the expected output shape.
Ground it in this repo: concrete file paths, real commands, the non-obvious
traps specific to this codebase — not generic advice.
Loading