ワンクリックで
nvim-pack-audit
Run interactive staged vim.pack.update() + security audit, then review findings before explicit approve/abort.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Run interactive staged vim.pack.update() + security audit, then review findings before explicit approve/abort.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Author a teach workspace's lessons/reference/index in markdown and build both the browsable HTML site and a phone-friendly EPUB from that one source.
A relentless interview that asks every frontier question at once, round by round.
Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker — edges as text in one file per ticket locally, or native blocking links on a real tracker.
Print the relevant text as raw copy-able markdown and copy it to the clipboard. Run `/markdown file` to copy it as a file reference instead, for pasting into Slack.
Move issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.
Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/PRD asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use when the user wants to review a branch, a PR, work-in-progress changes, or asks to "review since X".
| name | nvim-pack-audit |
| description | Run interactive staged vim.pack.update() + security audit, then review findings before explicit approve/abort. |
Use this skill when the user asks to update and audit Neovim vim.pack plugins.
Run:
core/ai/skills/nvim-pack-audit/scripts/start-audit
If it exits with no plugin updates applied, stop here.
Save the emitted state.json path from output (staging ready: ...).
audit_prompt_path, schema_path, old_lockfile, staged_lockfile, plugin_root,
audit_json, and audit_md from state.json.audit_prompt_path as its task, followed by a ## Runtime Context section with
mode: changed_only, old_lockfile_path: <old_lockfile>,
new_lockfile_path: <staged_lockfile>, plugin_root: <plugin_root>, and
audit_md_path: <audit_md>. The sub-agent must have read/write access to the repo and write its
JSON verdict to audit_json and its Markdown report to audit_md.core/ai/skills/nvim-pack-audit/scripts/record-audit <state-json>
This validates audit.json against the schema and writes summary.txt and commit-message.txt.
If validation fails, re-run the sub-agent.While waiting for promotion decision, support:
show summary
core/ai/skills/nvim-pack-audit/scripts/show-summary <state-json>show plugin <name>
core/ai/skills/nvim-pack-audit/scripts/show-plugin <state-json> <name>open audit report
core/ai/skills/nvim-pack-audit/scripts/open-audit-report <state-json>edit commit message
core/ai/skills/nvim-pack-audit/scripts/edit-commit-message <state-json> <output-file>Free-form follow-up questions are expected in this state.
approve
core/ai/skills/nvim-pack-audit/scripts/approve-update <state-json> [--commit-message-file <path>]abort
core/ai/skills/nvim-pack-audit/scripts/abort-update <state-json>done-auditing and review Q&A.abort and block.approve flow.core/ai/skills/nvim-pack-audit/scripts/cleanup <state-json>