Give your AI a pre-publish compliance check.
One MCP tool that checks any social content your agent writes against the current policies of 8 platforms — Meta, TikTok, LinkedIn, Google Ads, YouTube, X, Snapchat, Pinterest — beforeit ships. It returns the exact risky phrase, why it's risky, and a compliant rewrite.
Setup in 30 seconds
Add one server to your MCP client and paste your key — no SDK, no prompt engineering. Your agent discovers the tool automatically.
{
"mcpServers": {
"auditsocials-compliance": {
"command": "npx",
"args": ["-y", "auditsocials-compliance-mcp"],
"env": { "AUDITSOCIALS_API_KEY": "as_live_..." }
}
}
}What your agent can ask
One self-describing tool. Your agent sends a draft, gets back the exact risky phrases and how to fix them.
check_social_content_compliancecontentstringThe draft to check (required)platformsstring[]Any of the 8 — omit for allcontentTypeenumpost · caption · ad · video“guaranteed”State results vary; drop the guarantee.
“clinically proven”Remove unverifiable medical endorsement.
“cure anxiety”Use “may help with” instead of a cure claim.
Just ask — in plain language
No prompt engineering. Your agent picks the tool automatically. A few ways it comes up:
Check this caption against TikTok and Meta before I post it: “…”
Write 3 Instagram captions for my supplement — run each through a compliance check first.
Is this ad copy safe to run on Meta? Rewrite anything risky.
Review my launch tweet for all 8 platforms and flag policy risks.
Not just hard bans — the quiet risks too.
Most checks only tell you what's outright prohibited. Platforms rarely ban outright — they quietly reduce distribution, demonetize, or shadow-limit. Every finding carries a confidence label so your agent can act proportionately:
A specific, well-established policy term matched. High confidence — fix before publishing.
A borderline or semantic signal — reduced-distribution territory. Flagged, not overstated.
Findings are grounded in the same live policy engine behind the AuditSocials Policy Tracker and enforcement layer — never fabricated, and hedged when uncertain.
1 credit = 1 full-quality check — no watered-down free tier. No credit card.
Meta, TikTok, LinkedIn, Google Ads, YouTube, X, Snapchat, Pinterest + DSA signals.
Higher volume when you outgrow the free tier — usage-based.
Give your AI a pre-publish compliance pass
One MCP server, one free key, 30-second setup — no SDK and no prompt engineering. Your agent discovers the tool automatically and checks every draft against the current policies of all 8 platforms before it ships, returning the exact risky phrase and a compliant rewrite.