install
source · Clone the upstream repo
git clone https://github.com/ai-analyst-lab/ai-analyst
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/ai-analyst-lab/ai-analyst "$T" && mkdir -p ~/.claude/skills && cp -r "$T/.claude/skills/TEMPLATE" ~/.claude/skills/ai-analyst-lab-ai-analyst-template && rm -rf "$T"
manifest:
.claude/skills/TEMPLATE/skill.mdsource content
Skill: {{BLANK_1_SKILL_NAME}}
Purpose
{{BLANK_2_WHEN_TO_FIRE}}
When to Use
Fires automatically when the user asks Claude to do something that matches the trigger condition above.
Instructions
- Detect the trigger condition
- Execute your guardrail check
- If the check matters, print a clear, visible warning with "{{BLANK_3_SIGNATURE_PHRASE}}" as the first line
- Continue with the analysis, incorporating the warning into the output
Anti-Patterns
- Do not fire when the condition doesn't apply
- Do not block the analysis — just flag and continue