socialwtf-mcp

Scan Content with Sentinel Shield Safety Classifier

socialwtf_shield_scan

Read-onlyIdempotent

Run preliminary keyword screening on text, filenames, or media URLs to flag sensitive terms and fail closed on uninspected binary media with quarantine or age restriction.

Instructions

Run SocialWTF's in-house Sentinel Shield preliminary keyword screening on a text snippet, filename, or media URL. Distinguishes preliminary keyword screening (analysisMethod: 'preliminary_keyword_screen', mediaInspected: false, verifiedMediaSafety: false, confidence: null) from verified media content inspection, and fails closed ('quarantined' or 'age_restricted') on uninspected binary media URLs or sensitive terms. Works with or without an account and modifies no state.

Input Schema

Name	Required	Description	Default
`type`	No	Optional declared content type ('text', 'image', 'video', or 'audio'). Binary media types fail closed unless inspected by a verified media classifier.
`media`	Yes	Text snippet, caption, filename, or media URL to evaluate with Sentinel Shield (1-2000 characters).

Output Schema

Name	Required	Description
`tags`	No	Assigned safety tags.
`reason`	No	Human-readable explanation of the classification.
`scannedAt`	No	ISO-8601 timestamp of the scan.
`confidence`	No	Classifier confidence score (0-1), or null when heuristic keyword screening is used.
`isShielded`	No	True if the content is restricted or quarantined behind a safety shield.
`analysisMethod`	No	Screening method used (e.g. 'preliminary_keyword_screen').
`classification`	No	Safety classification ('safe', 'age_restricted', 'quarantined', or 'rejected').
`mediaInspected`	No	True only when actual image/video/audio bytes were inspected by a media classifier.
`publicFeedEligible`	No	Whether the item is eligible for public feeds.
`verifiedMediaSafety`	No	True only when media safety is verified by content inspection (false for keyword-only screening).

Schema Changelog

Changes observed during successful MCP inspections.

Addedv1.2.0Nov 5, 2026

TDQS

A4.2/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare readOnly/idempotent/non-destructive/openWorld; the description goes well beyond them by disclosing the fail-closed outcomes ('quarantined', 'age_restricted') for uninspected binary media or sensitive terms. It also pins the exact result semantics (analysisMethod 'preliminary_keyword_screen', mediaInspected false, verifiedMediaSafety false, confidence null) and notes no account is required.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, front-loaded with the verb and resource, with zero filler. The second sentence is dense—packing the result-field enumeration and fail-closed behavior into one clause—but each element carries information an agent needs.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists, so return values need no explanation here; the description still covers auth requirements, failure modes, and the boundary against verified inspection. For a two-parameter read-only scan tool, nothing an agent needs to call it correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so both parameters are already documented, and the description largely restates the media and type semantics rather than adding format or syntax detail. Its one genuine addition is confirming that binary media types fail closed unless inspected, which clarifies the behavior of the 'type' enum.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb ('Run ... preliminary keyword screening') and names the exact resource ('Sentinel Shield') plus accepted input kinds (text snippet, filename, media URL). It explicitly contrasts this with 'verified media content inspection', so an agent can tell the preliminary screen apart from the verified-classifier path without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies when this applies (preliminary screening of text/filename/URL, usable with or without an account) but never states a when-to-use/when-not rule or names a concrete alternative among the siblings such as socialwtf_sentinel_chat or socialwtf_sentinel_status. The distinction from 'verified media content inspection' is behavioral rather than routing guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.