postlake-inbox

postlake-inbox is a skill for Claude Code from PostLake/postlake-mcp. It costs 48 tokens per session (748 once invoked), scanned A, original, MIT.

A skill for reading and handling direct-message conversations from Facebook, Instagram, X, and Bluesky through one inbox. It covers listing threads, reading them, drafting replies, sending approved replies, and marking conversations read.

In plain words
What is it for?
Use it to review social inboxes, read conversation threads, prepare replies, send approved messages, and mark handled threads as read.
Why use it?
Checking several social platforms separately makes it easier to miss messages or use the wrong account. This provides one workflow and requires approval before sending a reply.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the postlake plugin — 8 skills, 2 MCP servers shipped together

Good fit Use it to review social inboxes, read conversation threads, prepare replies, send approved messages, and mark handled threads as read.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/postlake/postlake-mcp/postlake-inbox
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 PostLake/postlake-mcp --skill postlake-inbox
Clone the repo
git clone --depth 1 https://github.com/PostLake/postlake-mcp

Made for: Claude Code.

Or install postlake, the plugin that ships this one along with the rest of its 8 skills, 2 MCP servers.

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.

agentmods badge for postlake-inbox

README.md
[![agentmods](https://agentmods.dev/badge/skills/postlake/postlake-mcp/postlake-inbox/github.svg)](https://agentmods.dev/skills/postlake/postlake-mcp/postlake-inbox)
Your own site
<a href="https://agentmods.dev/skills/postlake/postlake-mcp/postlake-inbox"><img src="https://agentmods.dev/badge/skills/postlake/postlake-mcp/postlake-inbox/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.

agentmods 80×15 button for postlake-inbox

Your own site · 80×15
<a href="https://agentmods.dev/skills/postlake/postlake-mcp/postlake-inbox"><img src="https://agentmods.dev/badge/skills/postlake/postlake-mcp/postlake-inbox.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 748 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00048 $0.00748
Opus 5 $0.00024 $0.00374
Sonnet 5 $0.00010 $0.00150
Haiku 4.5 $0.00005 $0.00075

Measured 2d ago against content hash c24ad9affc36, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

postlake-inbox 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 2d 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.

skills/postlake-inbox/SKILL.md · 70 lines

How it starts

The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PostLake: unified inbox

Use PostLake's normalized inbox instead of calling four network APIs.

Start safely

  1. Call list_social_accounts to identify the connection.
  2. Call get_platform_capabilities if you do not know whether that connection supports messaging.
  3. Call list_conversations, optionally filtered by account.
  4. Check problems before claiming there are no messages.
  5. Call read_conversation with both the conversation id and account id.
  6. Draft a concise reply. Do not send until the user has asked you to send it.
  7. Call send_message, then mark_conversation_read after the thread is handled.

Conversation ids are scoped to one connected account. Never guess the account or reuse a conversation id with another connection.

REST equivalents

Action REST endpoint
List conversations GET /v1/conversations
Find or open a thread POST /v1/conversations with account and handle
Read messages GET /v1/conversations/{id}/messages?account=acc_...
Mark read POST /v1/conversations/{id}/read
Send a message POST /v1/conversations/{id}/messages

All REST requests use Authorization: Bearer $POSTLAKE_API_KEY. Reading, marking, and sending require the connected account id because a conversation id has meaning only within that connection.

What the response means

  • fromMe identifies which side sent a message. Do not infer this from handles.
  • content can describe an attachment, shared media, or an unsupported provider payload. Do not claim that an empty text means the message itself was empty.
  • The optional shape is { "kind": "attachment" | "shared_media" | "unsupported", "label": "...", "url": "..." | null }.
  • A problems entry means PostLake could not read a network. Report it. An empty items array without a problem means the network answered with no threads.
  • Cursors are opaque. Pass them back exactly as returned.

Network rules

  • Facebook and Instagram can produce message.received webhooks.
  • X and Bluesky require polling with list_conversations.
  • Sending a direct message on X costs 6 credits. Messaging on the other currently supported inbox networks does not spend credits.
  • Meta normal replies must be within 24 hours of the person's last message.
  • Never send humanAgent: true from an autonomous workflow. It asserts that a person wrote the reply, and Meta can penalize the connected account when that assertion is false. MCP deliberately does not expose this override.
  • Bluesky app passwords need direct-message access enabled when created.
  • Facebook and Instagram messaging may be unavailable until Meta grants the required permission to the connected business.

Read the full file on GitHub · 70 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.

  1. 2d ago First seen · 70 lines · 48 tokens per session scan A c24ad9affc36

Subscribe to this mod's changes

postlake-inbox is a skill published in the GitHub repository PostLake/postlake-mcp (0 stars, last pushed yesterday), licensed MIT. It adds 48 tokens to every session and 748 once invoked, about $0.0002 per session on Opus 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 2026-09-07.