Audit BMAD source files for file-reference convention violations using parallel Haiku subagents. Use when users requests an "audit file references" for a skill, workflow or task.
日本語の概要は準備中です。原文の説明を表示しています。
Generate social media announcements for Discord, Twitter/X, Facebook, and LinkedIn from the latest changelog entry. Use when user asks to 'create release announcement' or 'create social posts' or share changelog updates.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
Generate engaging social media announcements from changelog entries. This skill uses progressive disclosure: identify which platform(s) the user wants, then read only that platform's guide in references/.
Write each platform's announcement to {project-root}/.social/{social-type}/post.md, where {social-type} is one of discord, twitter, facebook, linkedin. Create the folders as needed; overwrite the platform's post.md to refresh it.
Look for .claude-plugin/marketplace.json in the project root. If found, read it to understand what plugins exist and their versions.
Single plugin: The announcement is for that plugin specifically. Use its name and version in all posts (e.g., "bmad-utility-skills v1.2.0" not just "BMad v1.2.0").
Multiple plugins: Ask the user which plugin(s) they're announcing. Each plugin should be named specifically in the posts.
No marketplace.json: Fall back to repo name and git tags.
The user may want to announce across multiple changelogs at once (e.g., releases in two repos the same week). If the user provides multiple changelog paths or mentions multiple repos:
Read ./CHANGELOG.md and extract the latest version entry. The changelog may use either format:
Standard format:
## [VERSION]
### 🎁 Features
* **Title** — Description
Multi-plugin format:
## plugin-name - vX.Y.Z - [Date]
* Change description
Parse:
6.0.0-Beta.5)Use git log to find contributors since the previous version. Get commits between the current version tag and the previous one:
# Find the previous version tag first
git tag --sort=-version:refname | head -5
# Get commits between versions with PR numbers and authors
git log <previous-tag>..<current-tag> --pretty=format:"%h|%s|%an" --grep="#"
Extract PR numbers from commit messages that contain # followed by digits. Compile unique contributors.
The difference between a flat release note and a post that sizzles is knowing what to put top-of-fold. This is a conversation to craft the headline with the user — it is not a dump of the full post in chat.
Only once the headline and angle are agreed do you move to generation — every platform post then leads with that sizzle.
Ask the user which platform(s) they want — discord, twitter, facebook, linkedin, or all. For each chosen platform, read its guide and follow it exactly. Load only the guide(s) for the platform(s) in play:
| Platform | Guide | Notes |
|---|---|---|
| Discord | references/discord.md | Hard 2,000-character limit |
| Twitter/X | references/twitter.md | Series of standalone, schedulable tweets |
references/facebook.md | Article-style thought piece | |
references/linkedin.md | Major/minor releases with significant features |
Each guide carries that platform's rules, template, and a pointer to its example.
Before you draft, read references/writing-quality.md and apply its Tier 1 (Universal AI tells) — banned vocabulary, "it's not X, it's Y", honesty preambles, weapons imagery, stakes inflation. Do NOT apply Tier 2 (long-form prose mechanics): social copy keeps its punch — fragments, short lines, emoji, and rule-of-three are on-format here, not errors.
Before finalizing, run a quality pass with a subagent (Task tool, subagent_type=general-purpose). Hand it the drafted post(s) and references/writing-quality.md, and instruct it to apply Tier 1 only — hunt every Universal AI tell (banned vocabulary, metanoia "it's not X, it's Y", hypophora, honesty preambles, weapons imagery, stakes inflation, sycophancy residue) and return concrete fixes. Tell it explicitly NOT to apply Tier 2 — fragments, short punchy lines, and rule-of-three are correct for social and must survive the pass. Apply the fixes.
Then write each refined result to {project-root}/.social/{platform}/post.md (for combined multi-plugin announcements, weave all plugins into that platform's single post.md), present a short preview per platform in the chat (include Discord's character count), and offer to adjust emphasis, tone, or content.
Include:
Skip/Minimize:
Emphasize:
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Audit BMAD source files for file-reference convention violations using parallel Haiku subagents. Use when users requests an "audit file references" for a skill, workflow or task.
日本語の概要は準備中です。原文の説明を表示しています。
Analyzes changes since last release and drafts them directly into the project's CHANGELOG.md. Use when users requests 'update the changelog' or 'prepare changelog release notes'
日本語の概要は準備中です。原文の説明を表示しています。
Create, update, fix, or refine documentation using Diataxis framework and BMad Method style guide. Use when user asks to 'create a doc', 'update docs', 'fix docs style', 'refine docs', or 'improve docs writing'.
日本語の概要は準備中です。原文の説明を表示しています。
Review translated documentation for content fidelity — detect injected, off-topic, or unauthorized content by comparing against the English source. Use when reviewing translation PRs or translated docs.
日本語の概要は準備中です。原文の説明を表示しています。
Orchestrate HITL triage of review findings using parallel agents. Use when the user says 'triage these findings' or 'run findings triage' or has a batch of review findings to process.
日本語の概要は準備中です。原文の説明を表示しています。
Analyze all github issues. Use when the user says 'triage the github issues' or 'analyze open github issues'.
日本語の概要は準備中です。原文の説明を表示しています。