Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add azeemkafridi/bulkpublish-api --skill comment-repliesgit clone --depth 1 https://github.com/azeemkafridi/bulkpublish-apiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/azeemkafridi/bulkpublish-api/comment-replies)<a href="https://agentmods.dev/skills/azeemkafridi/bulkpublish-api/comment-replies"><img src="https://agentmods.dev/badge/skills/azeemkafridi/bulkpublish-api/comment-replies/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/azeemkafridi/bulkpublish-api/comment-replies"><img src="https://agentmods.dev/badge/skills/azeemkafridi/bulkpublish-api/comment-replies.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00059 | $0.00647 |
| Opus 5.5 | $0.00024 | $0.00259 |
| Sonnet 5.5 | $0.00012 | $0.00129 |
| Haiku 4.5 | $0.00006 | $0.00065 |
Grade A, and why
comment-replies scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Comment Replies
Work through the comments under the user's own published posts: read them, group them, draft replies and act on the ones the user approves.
The rule that matters most
A reply is public and posted immediately, and a deleted comment cannot be restored. Before calling reply_to_comment, or moderate_comment with action: "delete":
- Show the exact reply text, or the comment you are about to delete.
- Name the post, the network and the account it happens on.
- Wait for an explicit yes for that one item. Approval of one is not approval of the next.
Liking, hiding and unhiding can be undone, but still say what you are about to do before doing it. Never act on a whole list at once.
Tools
list_postswithstatus: "published": find the post the user means.list_post_comments: the comments on that post, per channel, with each comment's id and thepostPlatformIdto answer it on. It also says when a channel cannot be replied to and why.reply_to_comment: posts a public reply (up to 2,000 characters) under one comment, using the post id, the channel'spostPlatformIdand the comment's id. To answer a reply, pass the thread's top comment asrootCommentId.moderate_comment:like,unlike,hide,unhideordelete. Only the actions a channel supports are listed for it inlist_post_comments; do not offer the others.
Recipe
- Find the post with
list_posts, then calllist_post_comments. - Group the comments: questions, praise, complaints, spam.
- Draft replies for the questions and complaints. Stick to what the post and the user say; do not invent facts.
- Confirm each reply, then
reply_to_comment. - Propose hiding obvious spam. Delete only when the user explicitly asks for that comment to be deleted.
When something is refused
Explain it in plain words and say what the user can do.
- Plan. Replying to and moderating comments is on every plan except Free. If a call is refused for the plan, say so and suggest upgrading. Reading comments works on every plan.
- Reconnect needed. Ask the user to reconnect that channel from the Channels page, then try again.
- Not available on that network. Some networks do not allow replies or moderation through BulkPublish. Tell the user to act on the network itself.
- Role. Deleting a comment needs an owner, admin or approver on the workspace.
- Too fast. Replies are limited to 10 a minute. Wait and continue.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 44 lines · 59 tokens per session scan A f1b512018f0c
comment-replies is a skill published in the GitHub repository azeemkafridi/bulkpublish-api (3 stars, last pushed yesterday), licensed MIT. It adds 59 tokens to every session and 647 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-21.
Other skills, from other repositories
zernio
Schedule posts, manage inbox, broadcasts, sequences, and automations across 14 platforms from the CLI.
schedule-social-post
Draft, schedule or publish social media posts to Instagram, TikTok, YouTube, Facebook, X, LinkedIn, Threads and Bluesky through the hosted Socialync MCP server, with the user's approval on every post. Use when the user wants to post, cross-post, publish now, schedule one post or a batch, send a video to TikTok, Reels…
blog-to-a-week-of-posts
Turn one blog post, newsletter, transcript or long article into a week of LinkedIn and X posts (plus Threads and Bluesky if the user wants), each one standing on its own, scheduled through Socialync with the user's approval on every post. Use when the user shares a URL or pastes long-form text and asks to repurpose…
film-once-week-of-shorts
Turn one filming session into a week of short-form video on TikTok, Instagram Reels and YouTube Shorts, scheduled through Socialync with the user's approval on every post. Use when the user has batch-filmed several clips and wants them spread across the week, or says "schedule my videos for the week", "post one clip a…
weekly-what-worked
Run a weekly review of the user's social media from their Socialync analytics, showing which platform grew, which posts did best and worst and what they have in common, which posts failed, and where next week's calendar has gaps, then propose next week's posts. Use when the user asks what worked, how last week went…
social-queue-analytics
Inspect and manage the Socialync calendar and post history. Use when the user asks what is scheduled, what already posted, why a post failed, or wants to cancel, move or publish early a scheduled Socialync post. For a weekly performance review or "what worked", use the weekly-what-worked skill.