| name | skillbuilder |
| description | Build flawless Claude Code skills. Studies existing skills as reference, ensures correct format, and pushes for genuine intelligence — skills that exploit something specific about how Claude works. SKIP for one-off scripts, prompts, or task helpers. |
| user-invocable | true |
| allowed-tools | ["Read","Write","Glob","Grep","Bash"] |
| effort | high |
/skillbuilder
Build Claude Code skills that are genuinely intelligent — not just format-valid, but skills that exploit something specific about how Claude works.
Reference standard: the skills in this repo. Study them before drafting.
| Skill | What it exploits |
|---|
/stochastic | LLM variance — poll N runs, aggregate consensus/divergences/outliers |
/model-chat | Disagreement — multi-agent debate surfaces assumptions a single pass misses |
/fanout | Parallelism — independent researchers cover more ground, Opus synthesizes |
/autoresearch | Measurement loops — read code, mutate, measure, keep-or-revert, repeat |
Every good skill exploits something. The question isn't "which of 8 buckets" — it's "what does this skill do that a raw prompt can't?"
Quality bar
A skill is worth building if:
- Reusable — works across many tasks, not a one-shot
- Exploits something — LLM variance, parallelism, measurement, memory, gates, whatever
- Beats a prompt — materially changes Claude's execution, not just vocabulary
- Passes the removal test — delete the skill, output measurably degrades
If the answer to "what does this exploit" is "nothing, it's just instructions" — that's a prompt, not a skill.
Anti-patterns
- Task helper. "Skill that writes commit messages." That's a prompt. "Skill that spawns 3 agents to review, vote, kill weak verbs" — that's a skill.
- Best-practice list. Documentation isn't a skill. A checklist that gates SHIP/NOT SHIP is.
- Prompt-equivalent. If a one-shot prompt produces the same output, don't build a skill.
- Single-use. "Skill to set up my repo." That's a script.
- Decoration. Skill exists to feel useful but doesn't change execution.
- Token bloat. SKILL.md > 250 lines of best-practice advice Claude would do anyway.
- Generic naming.
/helper, /assistant, /improve-X. Names should name what it does or what it exploits.
Operations
craft — build a new skill
Input: domain or problem. Output: working skills/{name}/SKILL.md.
Flow:
- Study references. Read existing skills in this repo. What patterns apply?
- Identify the exploit. What will this skill do that a raw prompt can't? If nothing — stop, recommend a prompt instead.
- Draft the skill following the template below.
- Write to disk.
skills/{name}/SKILL.md
- Report. Path, what it exploits, restart required.
audit — check an existing skill
Input: skill name or path. Output: lands · needs work · kill it.
Flow:
- Read the SKILL.md.
- Identify what it exploits. Implicit or missing → needs work.
- Run anti-patterns 1-7. Any hit → needs work or kill.
- Removal test: would Claude's output degrade without this skill?
- Verdict + one move.
kill — diagnose a failing skill
Input: skill name + symptom. Output: fix or kill.
Flow:
- What was it supposed to exploit? Has it drifted?
- Anti-pattern check.
- One move: sharpen, narrow, or kill.
Skill template
---
name: {kebab-case}
description: "TRIGGER when {specific signal}. {What it does in one phrase}. SKIP for {anti-domain}."
user-invocable: true
allowed-tools:
- {minimum needed}
effort: {default | high}
---
{What this skill does. What it exploits — why it beats a raw prompt.}
{If applicable: similar skills, shipped products, or patterns this builds on.}
{The mechanism. What Claude does differently when this skill is loaded.}
{Domain-specific failure modes to avoid.}
{What the skill produces. Format, structure, deliverable.}
Frontmatter spec
| Field | Required | Notes |
|---|
name | yes | kebab-case, matches folder name |
description | yes | TRIGGER when... SKIP for... — this is how Claude decides to load it |
user-invocable | yes | true if user can call it directly |
allowed-tools | yes | minimum needed — don't over-grant |
effort | yes | default (single-pass) or high (spawns agents, long-running) |
Allowed-tools — minimum needed
Read — almost always
Agent — if it spawns sub-agents
Bash Glob Grep — for codebase work
Write — only if it persists state
WebSearch WebFetch — only if it does live research
Don't grant tools the skill doesn't use.
Naming
- Mechanism-named —
/stochastic, /fanout (names the how)
- Domain-named —
/autoresearch, /skillbuilder (names the what)
- Avoid —
/helper, /assistant, /better-X, /improve-Y
After writing
- Tell user to restart Claude Code
- Test: invoke in fresh conversation, verify trigger matches
- Check: does output beat a raw prompt?
Voice
- Lead with what it exploits
- Every claim cites a reference (existing skill or shipped product)
- If it's a prompt in disguise, say so
- Short sentences, specific mechanisms
- One skill per run