菜单
Use when grilling should be grounded primarily in project artifacts. Loads `/grill`, checks project vocabulary and prior decisions first, and spawns bounded explorers when the answer should come from docs.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Load for large human-facing outputs: reports, docs, multi-section explanations, artifacts. Choose the right form for each beat (prose, diagram, table, mockup) and put the answer first, depth behind it. If the medium allows, use progressive disclosure.
Use when deciding where knowledge goes or reading/writing durable docs: AGENTS.md, .context/, KB, docs/, and work directories.
Anchor work in a work item: artifacts on disk, not in context.
Investigate and engage concurrently: don't block on evidence before surfacing your read.
Use when exploring or changing the codebase: read AGENTS.md first, use .context/CONTEXT.md for detail, keep intent docs succinct.
Use when recovering decisions or context from conversation history: session log navigation, transcript delegation.
| name | grill-with-docs |
| type | reference |
| description | Use when grilling should be grounded primarily in project artifacts. Loads `/grill`, checks project vocabulary and prior decisions first, and spawns bounded explorers when the answer should come from docs. |
| model-invocable | true |
Load /grill if it isn't already loaded.
Load /knowledge-layers if it isn't already loaded.
Use this when the challenge should be grounded primarily in project artifacts.
Check repo docs, existing prompts, prior decisions, active work artifacts, and project vocabulary before reaching for external sources.
When the answer should come from project materials beyond what is already in
context, spawn a bounded @explorer instead of bulk-reading the tree yourself.
Use @session-miner only when the missing context lives in transcripts rather
than artifacts.
If the conversation produces durable terminology or decisions, update the right documentation surface.
