What is manage-comments?

A comment-management skill for PostFast that reads comments on supported social posts and prepares replies, private Instagram replies, or deletions for review. It shows each action before sending it.

Use it to find unanswered comments, draft replies, send approved responses, issue Instagram private replies, or hide comments where supported.

It gives you one review step before public replies, private replies, or deletions happen.

manage-comments — Skill for Claude Code by peturgeorgievv-factory

Install

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.

Any agent

npx skills add peturgeorgievv-factory/postfast-mcp --skill manage-comments

Clone the repo

git clone --depth 1 https://github.com/peturgeorgievv-factory/postfast-mcp

Made for: Claude Code.

Or install postfast, the plugin that ships this one along with the rest of its 5 skills, 1 MCP server.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/peturgeorgievv-factory/postfast-mcp/manage-comments/github.svg)](https://agentmods.dev/skills/peturgeorgievv-factory/postfast-mcp/manage-comments)

Your own site

<a href="https://agentmods.dev/skills/peturgeorgievv-factory/postfast-mcp/manage-comments"><img src="https://agentmods.dev/badge/skills/peturgeorgievv-factory/postfast-mcp/manage-comments/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/peturgeorgievv-factory/postfast-mcp/manage-comments"><img src="https://agentmods.dev/badge/skills/peturgeorgievv-factory/postfast-mcp/manage-comments.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 86 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,630 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00086	$0.01630
Opus 5.5	$0.00034	$0.00652
Sonnet 5.5	$0.00017	$0.00326
Haiku 4.5	$0.00009	$0.00163

Measured yesterday against content hash 0ba44019d838, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

manage-comments 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 yesterday.

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.

plugins/postfast/skills/manage-comments/SKILL.md · 65 lines

How it starts

The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Answer comments with PostFast

The PostFast inbox holds comments on the user's own posts on TikTok, Instagram, Facebook Pages and Threads. It is a comments inbox: never describe it as direct messages. Each account's inboxCapable flag in list_accounts says whether its comments reach the inbox today, and comments arrive only from the time the account was connected; there is no older history.

Rules

Nothing is sent or deleted without a yes to the exact preview. Each action has its own preview tool, which sends nothing and returns a preview: preview_comment_reply for a public reply, preview_private_reply for an Instagram private reply and preview_comment_deletion for a deletion. Show the user the preview, wait for their yes in the conversation, and only then call the tool that carries it out: reply_to_comment, send_private_reply or delete_comment. This holds even when the user dictated the reply word for word: they still see the preview and say yes before it is sent. Never preview and carry out in the same turn.
Carry out exactly what was previewed. Pass the confirmToken from the preview with the same itemId and text, and the same workspaceId only if you passed one to the preview. If the user changes the text, preview it again and show the new preview.
The conversation decides whether and how long a reply can be. Read canReply, maxReplyLength, windowState and disabledReason on the conversation and keep to them. Don't assume a network's rules. When canReply is false, tell the user why, using disabledReason, instead of trying.
If preview_comment_reply, preview_private_reply or preview_comment_deletion is not among your tools, that action (a public reply, a private reply or a deletion) isn't available on this connection. Tell the user and suggest doing it in PostFast instead. Reading, hiding, restoring and sorting comments still work.

Find comments to answer

If the user has more than one workspace (list_workspaces), ask which one and pass its id as workspaceId on every call.
get_inbox_unread_count gives the total of unread comments.
list_inbox_conversations lists threads, newest activity first, one per post. Filter with unreadOnly, statuses (OPEN, SNOOZED or CLOSED), platforms or socialMediaIds, and page with page.
list_inbox_items with the conversation's id as conversationId lists its comments and replies, oldest first (order: "DESC" for newest first). Items have direction (INBOUND from people, OUTBOUND from the account), state (VISIBLE, HIDDEN or DELETED), the author, and the text.
A comment is unanswered when it is INBOUND and VISIBLE and no OUTBOUND item replies to it; a reply's parentExternalItemId is the externalItemId of the comment it answers.
Show each comment with its post (the caption, and the link from postPreview when there is one), the author, the text and when it was written, and say whether a reply is possible.
After you have shown the user a thread, call mark_inbox_conversation_read for it.

Read the full file on GitHub · 65 lines

Changes

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.

yesterday Changed · +2 lines 0ba44019d838
25d ago Changed 072094a4168c
1mo ago First seen · 63 lines · 86 tokens per session scan A 42ed4fa261ab

Subscribe to this mod's changes

manage-comments is a skill published in the GitHub repository peturgeorgievv-factory/postfast-mcp (7 stars, last pushed 2d ago), licensed MIT. It adds 86 tokens to every session and 1,630 once invoked, about $0.0003 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 2026-09-27.

Marketing and SEO ai claude mcp model-context-protocol scheduling social-media

gingiris-twitter-agent-ops

Twitter/X Agent Operations SOP v2 — Rebuilt from real dual-account operations. Real data: • CN account: grew from 1,150 → 3,679 followers (14 months), 1 tweet/day, OAuth1 direct post • EN account: Buffer API • Real incidents: cron silent-failed for 2 weeks unnoticed, analytics data off by 17×, thread performance worse…

Gingiris-1031/gingiris-skills · 360 tokens

zernio

Schedule posts, manage inbox, broadcasts, sequences, and automations across 14 platforms from the CLI.

zernio-dev/zernio-cli · 24 tokens

rhylthyme

Schedule multi-step procedures that a person carries out against a clock, with parallel tracks, shared equipment and steps that end when the person says so. Use when turning a lab protocol, recipe, event run sheet or workout into a timed plan; when several procedures must share one instrument (one centrifuge, one…

rhylthyme/rhylthyme-mcp · 167 tokens

publish-a-social-post

Compose, preview and publish a social media post with Ravenpost — Instagram, TikTok, X (Twitter), LinkedIn, Threads, Bluesky, YouTube, Pinterest and Telegram. Use whenever the user wants to post, tweet, share, cross-post, schedule, queue or draft something to their connected social accounts ("post this on Instagram…

target1m/ravenpost-mcp · 133 tokens

manage-scheduled-posts

Find, edit, reschedule, cancel or delete posts already in Ravenpost, and check how a published post actually landed on each account. Use when the user asks what is scheduled, wants to change or move a post ("move Friday's post to Monday", "fix the typo in the draft", "cancel the TikTok one"), asks whether a post went…

target1m/ravenpost-mcp · 108 tokens

prepare-post-media

Get images and video into Ravenpost's media library and into the right shape for each network — upload from a URL, a local path or a direct presigned upload, reformat to feed, square, landscape, story or reel canvases, and cut one image into several platform sizes at once. Use when a post needs a picture or video…

target1m/ravenpost-mcp · 110 tokens