ScriptGrain is an MCP server published by com.scriptgrain, hosted at mcp.scriptgrain.com. When last checked on 2026-11-22 it was working: it answers and lists its tools. It declares 6 tools, including start_session, create_profile, get_profile, check_voice_match, compare_voice.
Your writing voice measured across 45 attributes; drafts in that voice, scored against it.
2 connections through 5 shared values. Tap any node.
Start a free guest session with no account. Returns guest_token (pass it to every other tool), claim_url (give it to the person: it keeps the voice profile in a free ScriptGrain account, or deletes it), expires_at and limits (one profile, one extraction, no drafts). Unclaimed sessions are archived after 7 days and can still be claimed or deleted from the same link. At most 3 sessions a day from one network; reuse the token you have.
No parameters.
{
"name": "start_session",
"title": "Start a guest session",
"description": "Start a free guest session with no account. Returns guest_token (pass it to every other tool), claim_url (give it to the person: it keeps the voice profile in a free ScriptGrain account, or deletes it), expires_at and limits (one profile, one extraction, no drafts). Unclaimed sessions are archived after 7 days and can still be claimed or deleted from the same link. At most 3 sessions a day from one network; reuse the token you have.",
"inputSchema": {
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"properties": {}
},
"annotations": {
"title": "Start a guest session",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}None yet. Changes appear after a server's second daily check.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
Inferred from the host name and headers: what a server runs on, not who runs it.
Build the guest session's one voice profile from the person's own writing and start the stylometric extraction (45 measured attributes). Samples MUST be their own human-written prose. You can use writing you already hold for them (messages they typed to you, documents or emails they shared, posts of theirs you can read): their exact words, never your memory notes or summaries about them, and never text you or another AI wrote or rewrote. Tell the person which pieces you are sending before you call this; the result's samples_used lists each piece's first line and word count so you can show them. 1-20 samples, each under 5,000 words, 25,000 words in all; a sample needs 50+ words to count; 3+ samples totalling 3,000+ words gives a confident profile. Extraction usually takes under a minute: poll get_profile until status is 'complete'. The result carries claim_url for the person.
| Parameter | Type | Description |
|---|---|---|
| name | string | Profile name, e.g. "Sam's newsletter voice". |
| samples * | array | The person's own writing, one string per piece. |
| guest_token * | string | The guest_token from start_session. |
| english_variant | string | British (uk) or American (us) spelling. Defaults to uk. |
{
"name": "create_profile",
"title": "Create a voice profile",
"description": "Build the guest session's one voice profile from the person's own writing and start the stylometric extraction (45 measured attributes). Samples MUST be their own human-written prose. You can use writing you already hold for them (messages they typed to you, documents or emails they shared, posts of theirs you can read): their exact words, never your memory notes or summaries about them, and never text you or another AI wrote or rewrote. Tell the person which pieces you are sending before you call this; the result's samples_used lists each piece's first line and word count so you can show them. 1-20 samples, each under 5,000 words, 25,000 words in all; a sample needs 50+ words to count; 3+ samples totalling 3,000+ words gives a confident profile. Extraction usually takes under a minute: poll get_profile until status is 'complete'. The result carries claim_url for the person.",
"inputSchema": {
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"guest_token",
"samples"
],
"properties": {
"name": {
"type": "string",
"maxLength": 120,
"description": "Profile name, e.g. \"Sam's newsletter voice\"."
},
"samples": {
"type": "array",
"items": {
"type": "string",
"minLength": 1
},
"maxItems": 20,
"minItems": 1,
"description": "The person's own writing, one string per piece."
},
"guest_token": {
"type": "string",
"pattern": "^sg_guest_[0-9a-f]{48}$",
"description": "The guest_token from start_session."
},
"english_variant": {
"enum": [
"uk",
"us"
],
"type": "string",
"description": "British (uk) or American (us) spelling. Defaults to uk."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Create a voice profile",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}The guest profile's status and detail. While extraction runs it returns progress_percent; once status is 'complete' it includes the narrative voice description and all 45 measured attributes. Poll every ~10 seconds after create_profile.
| Parameter | Type | Description |
|---|---|---|
| profile_id * | string | The profile_id create_profile returned. |
| guest_token * | string | The guest_token from start_session. |
{
"name": "get_profile",
"title": "Get a voice profile",
"description": "The guest profile's status and detail. While extraction runs it returns progress_percent; once status is 'complete' it includes the narrative voice description and all 45 measured attributes. Poll every ~10 seconds after create_profile.",
"inputSchema": {
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"guest_token",
"profile_id"
],
"properties": {
"profile_id": {
"type": "string",
"format": "uuid",
"description": "The profile_id create_profile returned."
},
"guest_token": {
"type": "string",
"pattern": "^sg_guest_[0-9a-f]{48}$",
"description": "The guest_token from start_session."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get a voice profile",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Score any text (20+ words) against the guest profile by feature-delta measurement. Returns voice_match_score (0-1), the per-feature `deltas` behind it, notes rendered from them and score_context caveats. On-voice text typically scores 0.85-0.95; below 0.45 is clearly off voice. Cite the delta numbers, do not invent detail. Free; 5 checks an hour per network (shared with compare_voice).
| Parameter | Type | Description |
|---|---|---|
| text * | string | The text to score (20+ words, max ~50k chars). |
| profile_id * | string | A 'complete' guest profile id. |
| guest_token * | string | The guest_token from start_session. |
{
"name": "check_voice_match",
"title": "Score text against a voice",
"description": "Score any text (20+ words) against the guest profile by feature-delta measurement. Returns voice_match_score (0-1), the per-feature `deltas` behind it, notes rendered from them and score_context caveats. On-voice text typically scores 0.85-0.95; below 0.45 is clearly off voice. Cite the delta numbers, do not invent detail. Free; 5 checks an hour per network (shared with compare_voice).",
"inputSchema": {
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"guest_token",
"profile_id",
"text"
],
"properties": {
"text": {
"type": "string",
"minLength": 1,
"description": "The text to score (20+ words, max ~50k chars)."
},
"profile_id": {
"type": "string",
"format": "uuid",
"description": "A 'complete' guest profile id."
},
"guest_token": {
"type": "string",
"pattern": "^sg_guest_[0-9a-f]{48}$",
"description": "The guest_token from start_session."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Score text against a voice",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Off-voice check with no profile: score up to three pieces against a main piece. Each piece is pasted text (120+ words) or a public URL. Returns per-piece `score` (0-1: 0.8+ on voice, 0.55-0.8 drifting, below 0.55 off voice), `band`, `top_differences` and `top_matches` with the measured numbers, plus `notes`. Cite the numbers, do not invent detail. Free; 5 checks an hour per network (shared with check_voice_match).
| Parameter | Type | Description |
|---|---|---|
| main * | object | The reference piece. |
| others * | array | One to three pieces to score against the main piece (text or url each). |
| guest_token * | string | The guest_token from start_session. |
{
"name": "compare_voice",
"title": "Compare writing samples",
"description": "Off-voice check with no profile: score up to three pieces against a main piece. Each piece is pasted text (120+ words) or a public URL. Returns per-piece `score` (0-1: 0.8+ on voice, 0.55-0.8 drifting, below 0.55 off voice), `band`, `top_differences` and `top_matches` with the measured numbers, plus `notes`. Cite the numbers, do not invent detail. Free; 5 checks an hour per network (shared with check_voice_match).",
"inputSchema": {
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"guest_token",
"main",
"others"
],
"properties": {
"main": {
"type": "object",
"properties": {
"url": {
"type": "string",
"format": "uri",
"maxLength": 2048,
"description": "Or a public page URL."
},
"text": {
"type": "string",
"maxLength": 50000,
"description": "Pasted text, 120+ words."
},
"label": {
"type": "string",
"maxLength": 80
}
},
"description": "The reference piece.",
"additionalProperties": false
},
"others": {
"type": "array",
"items": {
"type": "object",
"properties": {
"url": {
"type": "string",
"format": "uri",
"maxLength": 2048
},
"text": {
"type": "string",
"maxLength": 50000
},
"label": {
"type": "string",
"maxLength": 80
}
},
"additionalProperties": false
},
"maxItems": 3,
"minItems": 1,
"description": "One to three pieces to score against the main piece (text or url each)."
},
"guest_token": {
"type": "string",
"pattern": "^sg_guest_[0-9a-f]{48}$",
"description": "The guest_token from start_session."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Compare writing samples",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}Whether the person has claimed the guest session yet: status (active, expired, archived or claimed), claimed, archived, expires_at and claim_url. Works after expiry too. Once claimed, the guest_token stops working; the profile is in the person's own account, reachable with their API key or https://mcp.scriptgrain.com/mcp.
| Parameter | Type | Description |
|---|---|---|
| guest_token * | string | The guest_token from start_session. |
{
"name": "claim_status",
"title": "Check a guest session's claim link",
"description": "Whether the person has claimed the guest session yet: status (active, expired, archived or claimed), claimed, archived, expires_at and claim_url. Works after expiry too. Once claimed, the guest_token stops working; the profile is in the person's own account, reachable with their API key or https://mcp.scriptgrain.com/mcp.",
"inputSchema": {
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"guest_token"
],
"properties": {
"guest_token": {
"type": "string",
"pattern": "^sg_guest_[0-9a-f]{48}$",
"description": "The guest_token from start_session."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Check a guest session's claim link",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}ScriptGrain measures how a person writes (45 stylometric attributes) and scores text against it. This endpoint needs no account: call start_session first and keep its guest_token for every other call. A guest session builds ONE voice profile and scores text against it; drafting and everything else need the person to claim the session (free) at claim_url, so give them that link. Build the profile from writing you already hold for the person: messages they have typed to you, documents or emails they shared, posts of theirs you can read. Use their exact words, not your memory notes or summaries about them, and leave out anything you or another AI wrote or rewrote for them. Tell the person which pieces you are sending, then call create_profile (3 to 10 pieces, 3,000+ words works best, 25,000 at most) and poll get_profile until status is 'complete'. check_voice_match then scores any text against the profile; compare_voice scores pieces against a main piece with no profile. Sessions are archived after 7 days unclaimed; the claim link still works after that.
Names, descriptions and schemas were written by the server and are shown as data.