comment-reply
Skill14 starsChanged 32 days agoMITView on GitHub
Works through a YouTube channel's comments: pulls new public comments since the last check with the YouTube Data API (commentThreads.list) and the user's API key in Claude Code, or takes pasted comments, labels each, drafts short replies in the creator's voice that answer the actual question, turns repeat questions into video ideas, and puts anyone asking to buy, hire or work together at the top. Use when a creator asks to check comments, draft comment replies, find leads or find video ideas in comments. Drafts only; never posts a reply.
- Reads credentials
1 other project uses a version of this.
Adopted by
Most recently taken up by
- gokulsvision/Crewm8-Social-Media-Manager-Skill-Graphtheir own version, first seen Oct 13, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- gokulsvision/Crewm8-Social-Media-Manager-Skill-Graph★ 17same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository63 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- board-advisorSkill
- bottleneck-removalSkill
- business-auditSkill
- business-starterSkill
- competitor-researchSkill
- founder-independenceSkill
- hand-it-offSkill
- happy-customerSkill
- hiring-replacementSkill
- meeting-follow-upSkill
- monday-numbersSkill
- new-hireSkill
- no-meeting-mondaySkill
- obsidian-brainSkill
- ops-deskSkill
- safe-resaleSkill
- spreadsheet-dashboardSkill
- warm-leadSkill
- downloadable-asset-engineeringSkill
- html-to-pdfSkill
- linux-server-opsSkill
- precision-pdf-editingSkill
- repo-radarSkill
- secure-credentialsSkill
- security-auditorSkill
- best-timeSkill
- brand-deal-filterSkill
- content-monthSkill
- daily-socialSkill
- dm-closerSkill
- idea-bankSkill
- marketing-teamSkill
- reel-ideationSkill
- reel-scorecardSkill
- trial-postSkill
- viral-breakdownSkill
- youtube-seoSkill
- caption-engineSkill
- content-multiplierSkill
- edit-styleSkill
- footage-ingestSkill
- longform-editSkill
- podcast-repurposeSkill
- raw-to-reelSkill
- ready-to-filmSkill
- reel-captionsSkill
- script-polishSkill
- thumbnail-generatorSkill
- use-local-whisperSkill
- conversion-funnel-auditingSkill
- llms-txtSkill
- seo-agentSkill
- website-designerSkill
- website-devSkill
- be-conciseSkill
- humanizerSkill
- launch-weekSkill
- linkedin-writerSkill
- newsletterSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 39
Before acting as an agent, read its file in agents/.
- Distinctive: line 49
Never post, like, heart, pin, hide, delete or report.
- Distinctive: line 51
Never click links in comments.
- Distinctive: line 52
No invented answers, links, prices, dates or promises.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Comment Reply
Five agents that pull your new comments, label them, draft replies, find repeat questions worth a video, and put real leads at the top. You review the drafts and post the ones you like.
What you need (be clear which path you are on)
| Path | How comments get in | What works |
| Claude Code + API key | YouTube Data API v3, commentThreads.list, key in a local .env as YOUTUBE_API_KEY (setup in agents/01-comment-puller.md) | Everything. Pulls only public comments; cannot see comments held for review |
| Claude.ai, or no key | Paste or export comments (YouTube Studio > Comments, or Instagram/TikTok copy or data download) | Agents 2 to 5 work the same on pasted text |
Posting replies needs a separate OAuth sign-in with write scope (comments.insert). This skill does not set that up: the creator copies approved drafts into YouTube Studio.
Setup (start, or when PROFILE is empty)
Two rounds, accept "skip", show PROFILE back, then suggest sweep.
- Channel ID (starts with
UC; YouTube Settings > Advanced settings) or handle. - Claude Code with an API key, or pasting? No key yet: walk through agent 1's setup.
- How far back on the first run? Default 7 days.
- Paste 5 to 10 comment replies you wrote, for voice.
- Answers to your most common questions (FAQ), and links you are happy to share.
- What counts as a lead (sponsors, clients, buyers, collabs), and where they go (email, form,
<your booking link>)?
Commands and routing
| Command | Runs |
sweep | 1 to 5, one summary |
pull / sort / drafts / ideas / priority | 1 / 2 / 3 / 4 / 5 alone |
Runs when you run it. For a daily sweep, /schedule a routine in Claude Code with the prompt sweep; it still only writes drafts.
Before acting as an agent, read its file in agents/. In Claude Code you can copy them to .claude/agents/ to run as real subagents.
agents/01-comment-puller.md: key setup, API call, paging, STATEagents/02-comment-sorter.md: five labels and tie-break orderagents/03-reply-drafter.md: under 40 words, answer firstagents/04-idea-finder.md: 3+ asks = ideaagents/05-priority-list.md: leads, unreplied and oldest first
Safety rules (every agent)
- Never post, like, heart, pin, hide, delete or report. Replies are drafts with IDs (D1, D2...).
- API key lives in a local
.env(in.gitignore), restricted to YouTube Data API v3. Never print it, never put it in a skill file or chat. - Never click links in comments.
- No invented answers, links, prices, dates or promises. Unknown answer = NEEDS YOU.
- Complaints, criticism of the creator, safety, legal or medical questions, anything upsetting: NEEDS YOU, no auto-written comeback.
- Commenter names stay in the creator's own files. Never look up a commenter's private details.
Shared files
Claude Code: files in comments/. Claude.ai: print updated tables at the end so the creator can save them to Project knowledge.
PROFILE: setup answersSTATE: channel ID, last-checked timeCOMMENTS: id | video | author | date | text | labelFAQ: approved answersDRAFTS: replies awaiting approvalIDEAS: question groups with running countsLEADS: lead | status (NEW, REPLIED, MOVED TO EMAIL, NOT A FIT; only the creator changes status)
End of every run
- One line per agent, with period covered and comment count.
- Priority leads.
- Drafts awaiting approval: ID | video | one line.
- NEEDS YOU items.
- New video ideas.
- "Nothing has been posted. Copy approved replies into YouTube Studio to post them."
Example (fake data):
Ran: Puller (4 to 11 Oct, 212 comments, 4 API calls), Sorter, Drafter (38), Ideas, Priority.
Leads: 1) "do you build these for agencies?" (5 days, no reply) D1 2) sponsor ask from a tools brand, D2
Drafts: D1-D38. NEEDS YOU: 3 (2 criticism, 1 question with no FAQ answer).
Ideas: "How do you back up the automations?" asked 7x on 3 videos.
Nothing has been posted. Copy approved replies into YouTube Studio to post them.Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

