add-skill
Add Skill
Create a new joelclaw skill. Skills are modular instruction sets that extend agent capabilities with specialized knowledge, workflows, or tool integrations.
Process
1. Create the skill directory
mkdir -p ~/Code/joelhooks/joelclaw/skills/<skill-name>
2. Write SKILL.md
Every skill needs a SKILL.md with frontmatter and instructions:
---
name: <skill-name>
displayName: <Human Readable Name>
description: <One-line description. This shows in the skill list and is used for trigger matching.>
version: 0.1.0
author: joel
tags:
- <relevant>
- <tags>
---
# <Skill Title>
<Instructions for the agent. Write for another Claude instance — include non-obvious procedural knowledge, domain-specific details, gotchas, and reusable patterns.>
## When to Use
<Trigger phrases and situations that should activate this skill.>
## Operations
<Step-by-step procedures, commands, API calls, etc.>
## Rules
<Constraints, safety boundaries, things to never do.>
3. Add references (optional)
For complex skills, add supporting files:
skills/<skill-name>/
├── SKILL.md # Required
├── references/ # Optional — detailed docs, examples
│ └── operations.md
├── scripts/ # Optional — helper scripts
└── assets/ # Optional — logos, templates
├── small-logo.svg # For Codex desktop
└── large-logo.png # For Codex desktop
4. Symlink to all consumer directories
ln -sf ~/Code/joelhooks/joelclaw/skills/<skill-name> ~/.pi/agent/skills/<skill-name>
ln -sf ~/Code/joelhooks/joelclaw/skills/<skill-name> ~/.agents/skills/<skill-name>
ln -sf ~/Code/joelhooks/joelclaw/skills/<skill-name> ~/.claude/skills/<skill-name>
5. Slog it
slog write --action configure --tool skills --detail "created <skill-name> skill: <what it does>" --reason "<why>"
6. Commit
The skills/ directory is sacred and fully git-tracked. Every skill must be committed.
cd ~/Code/joelhooks/joelclaw
git add skills/<skill-name>
git commit -m "feat(skills): add <skill-name> — <short description>"
Key Rules
- Repo is canonical:
~/Code/joelhooks/joelclaw/skills/is the source of truth. Home dirs symlink to it. - Directory name must match
namefield in SKILL.md frontmatter. Mismatch causes[Skill conflicts]warning on pi load. - Never copy skills — always symlink.
cat > symlinkwrites through and destroys the target. - External/third-party skill packs stay external (global install), not copied into repo unless intentionally curated.
- Pi extensions load at session startup only — new skills are available immediately (loaded on demand), but if you modify an existing skill mid-session, run
/reload. - One skill per concern — don't overload a skill with unrelated capabilities. Split into focused skills.
- Write for another agent — the consumer is another Claude instance, not Joel. Include what's non-obvious.
- Include trigger phrases in the description — this is how pi matches user requests to skills.
Updating Existing Skills
- Edit the SKILL.md (or references) in the repo copy
- Symlinks mean all consumers see the change immediately
- Slog the change
- Commit
Codex Desktop Metadata (optional)
For skills that should appear in Codex desktop:
skills/<skill-name>/
├── agents/
│ └── openai.yaml # Codex agent config
└── assets/
├── small-logo.svg
└── large-logo.png
More from joelhooks/joelclaw
cli-design
Design and build agent-first CLIs with HATEOAS JSON responses, context-protecting output, and self-documenting command trees. Use when creating new CLI tools, adding commands to existing CLIs (joelclaw, slog), or reviewing CLI design for agent-friendliness. Triggers on 'build a CLI', 'add a command', 'CLI design', 'agent-friendly output', or any task involving command-line tool creation.
129k8s
>-
88docker-sandbox
Create, manage, and execute agent tools (claude, codex) inside Docker sandboxes for isolated code execution. Use when running agent loops, spawning tool subprocesses, or any task requiring process isolation. Triggers on "sandbox", "isolated execution", "docker sandbox", "safe agent execution", or when working on agent loop infrastructure.
86joel-writing-style
Joel's writing voice and style guide for joelclaw.com content. Use when writing, editing, or reviewing any blog post, essay, book chapter, or prose content for joelclaw.com. Also use when asked to 'write like Joel,' 'match Joel's voice,' 'draft a post,' 'write content for the blog,' or 'review this for voice.' This skill captures Joel's specific writing patterns derived from ~90,000 words of published content spanning 2012–2026. Cross-reference with copy-editing and copywriting skills for marketing-specific copy.
81task-management
Manage Joel's task system in Todoist. Triggers on: 'add a task', 'create a todo', 'what's on my list', 'today's tasks', 'what do I need to do', 'remind me to', 'inbox', 'complete', 'mark done', 'weekly review', 'groom tasks', 'what's next', or when actionable items emerge from other work. Also triggers when Joel mentions something he needs to do in passing — capture it.
54skill-review
Audit and maintain the joelclaw skill inventory. Use when checking skill health, fixing broken symlinks, finding stale skills, or running the skill garden. Triggers: 'skill audit', 'check skills', 'stale skills', 'skill health', 'skill garden', 'broken skill', 'skill review', 'fix skills', 'garden skills', or any task involving skill inventory maintenance.
49