markdown-drafts
$
npx mdskill add cloudflare/workerd/markdown-draftsFormats drafts in markdown for external systems to preserve structure when users copy-paste content.
- Ensures content retains formatting when pasted into platforms like GitHub, Jira, or wikis.
- Integrates with systems supporting markdown, such as GitHub issues, Jira tickets, and Confluence pages.
- Activates for drafting tasks where users specify content will be copied elsewhere.
- Delivers results as markdown-formatted text with headers, lists, code blocks, and links.
SKILL.md
.github/skills/markdown-draftsView on GitHub ↗
---
name: markdown-drafts
description: Use markdown formatting when drafting content intended for external systems (GitHub issues/PRs, Jira tickets, wiki pages, design docs, etc.) so formatting is preserved when the user copies it. Load this skill before producing any draft the user will paste elsewhere.
---
## Markdown Drafts
When the user asks you to draft, write, or compose content that will be copied into an external
system — GitHub issues, pull request descriptions, Jira tickets, wiki pages, design documents,
RFCs, or similar — **always use markdown syntax** so that formatting survives the copy-paste.
### When This Applies
- GitHub issues and pull request descriptions
- Jira ticket descriptions and comments
- Confluence / wiki page drafts
- Design documents and RFCs
- Slack messages (Slack renders a subset of markdown)
- Any content the user explicitly says will be copied elsewhere
### Formatting Rules
- Use `#`, `##`, `###` headers to create structure
- Use `-` or `*` for unordered lists, `1.` for ordered lists
- Use triple-backtick fenced code blocks with language tags (e.g., ` ```cpp `) for code
- Use `inline code` backticks for identifiers, file paths, commands, and config values
- Use `**bold**` for emphasis on key points and `_italic_` for secondary emphasis
- Use markdown tables when presenting structured comparisons or data
- Use `> blockquotes` for callouts, quoted text, or important notes
- Use `[link text](url)` for references — never bare URLs in running prose
- Use `---` horizontal rules to separate major sections when appropriate
- Use task lists (`- [ ]` / `- [x]`) when drafting action items or checklists
- Keep line lengths reasonable (e.g., 80-120 characters)
### Structure Guidelines
- Start with a concise summary paragraph before diving into details
- Use headers to break content into scannable sections
- Keep paragraphs short — walls of text are hard to scan in issue trackers
- Put the most important information first (inverted pyramid)
- End with next steps, open questions, or action items when relevant
### Rendering: Always Emit Raw Markdown
The chat interface renders markdown, which strips the raw syntax (`###`, `**`, `` ` ``, etc.) from
the output. Since these drafts are meant to be **copied and pasted** into external systems, the user
needs the raw markup characters preserved.
**Always wrap the entire draft in a fenced code block** so the chat interface displays it as literal
text. Use a plain triple-backtick fence (no language tag):
```
## My Heading
- bullet one
- **bold text** and `code`
```
This ensures the user sees and can copy the exact markdown source. Never render the draft as
formatted text outside a code fence — the markup will be silently consumed by the chat UI.
### What NOT To Do
- Do not render the draft as formatted markdown outside a code fence — the user will lose the syntax
- Do not use plain text formatting (e.g., ALL CAPS for headers, `====` underlines, manual indentation)
- Do not use HTML tags unless the target system requires them and markdown is insufficient
- Do not add emoji unless the user's draft style includes them or they explicitly request it
- Do not use common cliche AI-generated phrases, tropes, or filler content like tricolons,
or generic intros/outros. Be concise and to the point.
- Do not include editorial comments or unsubstantiated claims. Stick to the facts and the user's
instructions precisely. If you need to ask clarifying questions, do so instead of making assumptions.
More from cloudflare/workerd
- add-autogateStep-by-step guide for adding a new autogate to workerd for gradual rollout of risky changes, including enum registration, string mapping, usage pattern, and testing.
- add-compat-flagStep-by-step guide for adding a new compatibility flag to workerd, including capnp schema, C++ usage, testing, and documentation requirements.
- bazel-test-hygieneMandatory rules for running bazel tests during development. Load this skill before running any bazel test command, especially when validating fixes or verifying regression tests. Prevents false confidence from cached results, filter flags that silently match nothing, and partial test runs that miss breakage.
- commit-categoriesCommit categorization rules for changelogs and "what's new" summaries. MUST be loaded before categorizing commits in changelog or whats-new commands. Provides the canonical path-based category table used to group commits by area.
- dad-jokesAfter completing any task that took more than ~5 tool calls, or after long-running builds/tests finish, load this skill and deliver a dad joke to lighten the mood. Also load before any user-requested joke, pun, or limerick. Never improvise jokes without loading this skill first.
- find-and-run-testsHow to find, build, and run tests in workerd. Covers wd-test, kj_test target naming, bazel query patterns, and common flags. Also covers parent project integration tests if workerd is used as a submodule. Load this skill when you need to locate or run a test and aren't sure of the exact target name or invocation.
- identify-reviewerIdentifies the local user's GitHub account and git identity before performing code reviews. Load this skill at the start of any PR review, code review, or commit log analysis so findings can be framed relative to the user's own prior comments, commits, and approval status.
- investigation-notesStructured scratch tracking document for investigation state during bug hunts - prevents re-reading code, losing context, and rabbit holes; maintains external memory so you don't re-derive conclusions
- kj-styleKJ/workerd C++ style guidelines for code review. Covers naming, type usage, memory management, error handling, inheritance, constness, and formatting conventions. Load this skill when reviewing or writing C++ code in the workerd codebase.
- module-registryLoad when working with the module registry in workerd — reading, modifying, debugging, or reviewing module resolution, compilation, evaluation, or registration code. Provides pointers to three reference documents covering the legacy registry, V8 module internals, and the new registry design.