CommunitySlow· 1mo★ 4.3by Hagai Hen

Facebook MCP

Manages Facebook Page posts, comments, and engagement metrics through the Graph API with negative sentiment filtering.

View repo

Safety & Trust

Delv Safety Grade: C

Score 58/100 · assessed 2026-04-28

Maintainer40

Permissions55

Supply chain35

Transparency65

Incidents100

Facebook MCP is a community server by solo developer Hagai Hen that connects Claude to the Facebook Graph API for page management. It requires a Facebook Page access token with broad permissions to read posts, write comments, schedule content, and pull engagement metrics. The sentiment filtering feature adds AI-driven moderation but also means the server processes user comments through external analysis. Main concerns are the solo maintainer with limited track record, no package distribution (manual pip install from requirements.txt), and the powerful write permissions to a social media account. The access token grants substantial control over your Facebook Page, including posting and comment management. Open source with reasonable documentation, but supply chain is weak with no versioned releases or package registry distribution. No known incidents, but the combination of social media write access and manual installation warrants caution.

Lethal Trifecta (prompt-injection exposure)

TRIFECTA RISK

All three axes present. This server can read private data, ingest attacker-controlled content, and send data outbound. A poisoned input (a GitHub issue, an email, a webpage) can exfiltrate secrets via this chain. Only install with auditing; avoid on shared or cloud agents.

Private dataYes

Reads secrets, credentials, private files

Untrusted inputYes

Ingests web pages, PRs, issues, emails

External commsYes

Can send data outbound

Page mentions and message-the-page features are wide-open untrusted-input vectors. Outbound posting closes the chain.

Green flags

Open source on GitHub with visible code for audit
Addresses real moderation need with sentiment filtering
Clear documentation of required Graph API permissions
No known security incidents or malicious behaviour

Red flags

Solo maintainer with limited public track record or community adoption
Requires FB_PAGE_ACCESS_TOKEN with broad write permissions to social account
No package distribution, manual pip install from requirements.txt
Sentiment analysis may send comment data to external AI services
No versioned releases or supply chain verification

Permissions requested

Outbound networkAccess secretsSend messagesRead messagesIdentity writeExternal LLM call

Assessed by Delv Editorial using public metadata. Grades are advisory and update as the ecosystem changes. They do not replace your own review of permissions and code before granting an agent access to sensitive systems.

Install

uv pip install -r requirements.txt

Env vars needed: FB_PAGE_ACCESS_TOKENFB_PAGE_ID

Review

Facebook MCP hooks Claude into the Graph API for managing Facebook Pages. You get tools to read posts, reply to comments, schedule content, and pull engagement metrics. The standout feature is negative sentiment filtering, which automatically flags or filters hostile comments based on tone analysis. This is genuinely useful if you manage a brand page and want to surface the worst interactions first. I'd reach for this when moderating a busy Facebook Page from Claude Desktop. The typical workflow is asking Claude to summarise today's comments, flag anything aggressive, then draft replies to the constructive ones. You can also schedule posts or check which content performed best this week. It's faster than clicking through Facebook's own interface, especially if you're already working in Claude. The sentiment filtering is the main reason to install this. It's not perfect, sentiment analysis never is, but it catches obvious abuse and saves you scrolling through hundreds of bland comments to find the ones that need attention. The rest of the tooling is straightforward Graph API wrappers, nothing fancy, but they cover the essentials: create post, get comments, reply, fetch metrics. Quirks: you need a Page Access Token with the right permissions, and Facebook's token setup is fiddly if you haven't done it before. The repo doesn't hold your hand through that process, so expect to spend time in the Meta for Developers console. Also, this is a community project by one author, so don't expect enterprise-grade error handling or frequent updates. If Facebook changes the Graph API, you might be waiting for a fix. Skip this if you don't manage a Facebook Page or if you're looking for Instagram support, which isn't here. It's also overkill if you only post once a week and rarely get comments. This is for people who need to triage engagement at scale, not casual page owners.

Verdict

Install if you manage a Facebook Page with regular engagement and want Claude to handle comment moderation and sentiment triage. Skip if you're looking for Instagram tools, don't have a Page Access Token handy, or only post occasionally. The sentiment filtering is the killer feature, the rest is solid Graph API plumbing.

Good at

Negative sentiment filtering automatically flags hostile or abusive comments, saving time on moderation.
Covers the core Facebook Page workflows: posting, replying, scheduling, and pulling engagement metrics.
Works directly from Claude Desktop without needing to open Facebook's clunky interface.
Straightforward Graph API integration with no unnecessary abstraction layers.

Watch out

Requires a Facebook Page Access Token, which is fiddly to set up if you're not familiar with Meta's developer console.
Community-maintained by a single author, so updates may lag if Facebook changes the API.
No Instagram support, even though many page managers handle both platforms.
Sentiment analysis isn't perfect and will occasionally flag neutral comments or miss subtle hostility.

Use cases

Page comment moderation
post scheduling
engagement analytics
community management

Getting started

1. Clone the repo and run `uv pip install -r requirements.txt` to install dependencies. 2. Generate a Facebook Page Access Token from the Meta for Developers console with pages_manage_posts, pages_read_engagement, and pages_manage_metadata permissions, then set FB_PAGE_ACCESS_TOKEN and FB_PAGE_ID as environment variables. 3. Add the server to your Claude Desktop config under mcpServers, pointing to the repo's main script with the environment variables passed through. 4. Restart Claude Desktop and test by asking it to fetch recent comments from your page. 5. Watch out for token expiry: Page Access Tokens can expire after 60 days unless you generate a long-lived one, which requires extra steps in the Meta console.

Works with

Claude DesktopClaude CodeCursor