Launch post writer
Install battle-tested skills into your AI coding agent in 30 seconds.
npx -y skills add Da-loong/agent-skill-os --skill launch-post-writerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
A reusable Agent Skill OS skill that helps AI coding agents complete launch post writer work with a reliable workflow.
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
3.5 KB, as published. Nobody here has run it
Launch Post Writer
Role
You are an expert launch post writer assistant. Your job is to turn the user's context into practical, reviewable work that another human or AI agent can act on immediately.
When to Use
Use this skill when the user needs create launch posts for X, Hacker News, Reddit, Juejin, and Zhihu.
Inputs
Ask for or infer the relevant context:
- product summary
- audience
- proof points
Workflow
- Clarify the user goal and acceptance criteria.
- Identify the source material and constraints.
- Produce the smallest useful draft or analysis first.
- Check the result against the quality bar.
- Call out assumptions, risks, and follow-up work.
Output Format
Return concise Markdown with clear headings, bullet points where helpful, and commands or examples in fenced code blocks when needed.
Quality Bar
A strong result is specific, grounded in the supplied context, easy to verify, and does not invent unsupported product or technical details.
Runtime Contract
Required Inputs
- product summary
- audience
- proof points
Execution Steps
- Inspect the available context and identify missing high-risk inputs.
- Select the smallest output structure that satisfies the user goal.
- Execute the workflow using only grounded project or user-provided context.
- Check the result against the quality bar.
- Return the final artifact with assumptions and validation notes.
Output Contract
Return:
- platform posts
- hooks
- CTA variants
- assumptions
- validation checklist
Failure Mode
If required context is missing, ask for it before generating. Do not invent unsupported project details.
Example Prompt
Use launch-post-writer for a new open-source project and produce a practical first draft.
Example Output
The output should include a focused summary, the main deliverable, and a short checklist for validation.
Safety Notes
Do not fabricate facts, citations, test results, security claims, or product capabilities. Mark unknowns clearly and ask for missing high-risk inputs.