一键导入
lint
Run the lint pipeline and fix issues with judgment. Don't sweep errors under the rug.
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。
菜单
Run the lint pipeline and fix issues with judgment. Don't sweep errors under the rug.
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
Build or refine harness context - skills, runbooks, or specs. Interview for shape, identify the right layer, study existing files for voice, draft, iterate.
Sign the Firefox Browser Bridge extension via Mozilla AMO
Record progress - update by default, complete a topic with "done", record everything with "full".
Orient a new session — load memories, announce on the roster, follow memory pointers, report what you see
Name and describe unnamed Claude Code sessions, one at a time with confirmation.
Draft a commit message for the current changes.
| name | lint |
| description | Run the lint pipeline and fix issues with judgment. Don't sweep errors under the rug. |
Read ${CLAUDE_PLUGIN_ROOT}/doc/ai/runbook/lint.md for invocation, the tool
pipeline, spec pointers, and architecture docs. What
follows here is how to hold the fixes.
Separate obvious fixes from judgment calls. Obvious fixes (unused imports, spacing, naming drift) - just fix them. Judgment calls - surface them. When the convention doesn't cover the case, or the fix isn't clear, present the finding and the options before changing anything.
Findings the linter can't resolve are decisions, not facts. "Cannot de-alias", "cannot fix", informational reports that survive a lint pass - each one is a question, not a conclusion. Present them one at a time. Don't batch them, don't declare them permanent, don't categorize them as acceptable without input.
Don't suppress errors with _. Only byte counts
get the blank identifier, never error returns. If the
error genuinely can't happen, explain why.
Don't add lint-ignore annotations. No //nolint,
no //lint:ignore. If the linter is wrong, the fix is
in the code or in the analyzer - not in a comment.
Don't invent conventions. When a lint finding reveals a gap - a pattern the conventions don't address - flag it. That's a decision, not a fix.