Q&A Answer Crafting
Draft and post authoritative answers on Q&A platforms that build expertise reputation and drive profile clicks
npx gtm-skills add drill/qa-answer-craftingWhat this drill teaches
Q&A Answer Crafting
This drill produces high-quality answers for Q&A platforms (Stack Overflow, Quora, Dev.to) that establish technical authority and drive profile visits. Every answer must be genuinely the best answer to the question -- not a vehicle for marketing.
Input
- A question from the answer queue (from
qa-question-discoverydrill) - Your product's expertise areas, code examples, and relevant content
- Platform-specific profile and reputation data
Steps
1. Deep-read the question
Using qa-platform-api-read, fetch the full question with all existing answers:
Stack Overflow:
GET /questions/QUESTION_ID?site=stackoverflow&filter=withbody
GET /questions/QUESTION_ID/answers?site=stackoverflow&filter=withbody&sort=votes
Dev.to:
GET /api/articles/ARTICLE_ID
GET /api/comments?a_id=ARTICLE_ID
Analyze:
- What is the actual problem? Strip away the poster's framing and identify the root technical issue.
- What have existing answers covered? Map each answer's approach and identify gaps.
- What is wrong with existing answers? Outdated library versions, deprecated APIs, security issues, incomplete solutions.
- What is your unique angle? A better approach, a simpler solution, a production-tested pattern, a common gotcha.
If existing answers fully and correctly solve the problem and you have nothing to add, skip the question. Move to the next item in the queue.
2. Select answer format
Choose based on the question type:
The Direct Solution -- for specific "how do I X?" questions:
- Lead with working code
- Explain what the code does and why
- Note edge cases and error handling
- Include library versions tested against
The Architecture Answer -- for design/approach questions:
- Present 2-3 approaches with tradeoffs
- Recommend one with clear reasoning
- Include a minimal code example for the recommended approach
- Link to further reading for each approach
The Debugging Walkthrough -- for "why is X happening?" questions:
- Explain the root cause first
- Show the fix
- Explain why the original code failed
- Suggest how to prevent this class of bug
The Comparison Answer -- for "X vs Y" questions:
- Structured comparison table
- Specific use-case recommendations
- Personal experience with each option (if genuine)
- Benchmarks or data if available
3. Draft the answer
Use ai-content-ghostwriting to generate a draft, then refine. The draft prompt should include:
- The full question text
- Existing answers (so the draft avoids repetition)
- Your specific expertise and examples
- The selected answer format
Formatting rules by platform:
Stack Overflow:
- Use Markdown: headers, code blocks with language hints, bullet points
- Code blocks must be syntactically valid and tested
- Lead with the solution, then explain. Stack Overflow readers scan for code first.
- Include version numbers: "Tested with Node.js 20.x and Express 4.x"
- Keep under 800 words for answers, 300 words for comments
- Never use emoji. Tone: direct, technical, precise.
Quora:
- Use Quora's rich text: bold key points, numbered lists
- More conversational tone than Stack Overflow
- Can include personal experience and opinion
- Lead with a direct 1-2 sentence answer, then elaborate
- 200-600 words is the sweet spot
- Include relevant images or diagrams if they clarify the answer
Dev.to:
- Full Markdown with code blocks, headers, images
- More narrative style -- Dev.to readers expect blog-post quality
- Include "TL;DR" at the top for long answers
- Can be 500-1500 words
- End with a discussion prompt ("What approach do you use?")
4. Code validation (Stack Overflow only)
Before posting any answer with code:
- Syntax check: Run the code through a linter for the relevant language
- Execution check: If possible, execute the code and verify it produces the claimed output
- Version check: Verify library versions mentioned are current (not deprecated)
- Security check: Ensure code does not demonstrate insecure patterns (SQL injection, eval(), hardcoded secrets)
If the code cannot be validated, add a caveat: "Note: this approach requires [specific version]. Check [docs link] for your version."
5. Apply self-promotion guardrails
Before posting, run this checklist:
- [ ] Does this answer stand on its own without any link to your product? It must.
- [ ] If including a link to your content, is it the single best resource for this specific question?
- [ ] Is the link to educational content (blog post, open-source repo, docs) and NOT a landing page, pricing page, or signup page?
- [ ] Have you included at least 2 non-self links for every 1 self-link?
- [ ] In your last 10 answers on this platform, have you linked to yourself fewer than 2 times?
Stack Overflow explicitly bans promotional content. Quora is more lenient but communities self-police. Dev.to allows "shameless plugs" only in dedicated threads.
Profile optimization (do this once, not per answer):
- Stack Overflow: Set your profile bio to include your expertise area and a link to your site. Answers drive profile views; the profile drives clicks.
- Quora: Fill out credentials for your expertise areas. Quora displays credentials next to your answers.
- Dev.to: Complete your bio with your expertise and website link.
6. Post the answer
Using qa-platform-api-write:
Stack Overflow:
POST /questions/QUESTION_ID/answers/add
body=HTML_ENCODED_ANSWER&site=stackoverflow
Dev.to: Post as an article response or comment depending on length and platform norms.
Quora: Draft the answer. Human action required: Post via Quora's web interface until API access is confirmed.
7. Track the answer
Using community-engagement-tracking, log:
{
"date": "2026-03-30",
"platform": "stackoverflow",
"question_id": "12345",
"question_url": "https://stackoverflow.com/questions/12345/...",
"answer_url": "https://stackoverflow.com/a/12346/...",
"answer_format": "direct-solution",
"topic": "rate-limiting",
"tags": ["node.js", "api"],
"included_link": false,
"word_count": 350,
"code_included": true,
"status": "posted"
}
8. Follow up at 24h and 72h
Using qa-platform-api-read, check each posted answer after 24 hours and 72 hours:
- Upvote count: Log it. Answers with 3+ upvotes in 24h are strong signals.
- Accepted: Did the OP accept your answer? Log it.
- Comments: Did anyone ask follow-up questions? Respond promptly -- engaged follow-ups build more reputation than the original answer.
- Competing answers: Did a better answer appear? If so, note what they did differently.
Update the activity log with performance data.
Output
- Posted answer on the Q&A platform
- Activity log entry with performance metrics (updated at 24h and 72h)
- Follow-up responses to comments
Triggers
- Triggered by
qa-question-discoverydrill output (answer queue) - Target: 2-3 answers per day during active levels
- Skip days when no high-quality opportunities exist -- no filler answers