Skip to main content
Glama

Use the user's saved sign-ins

writ_personas

The user's own accounts on websites (personas): how Writ signs in as them. No password passes through this tool or the conversation, and its answers never ask for one: credentials are typed only into a Writ window. action='list' domain= shows the personas usable there; with none it answers persona_needed: a tell_user for the user, a create_url (a small Writ window with only the sign-in form, pre-filled for the site) and a link_id. action='wait' link_id= is one held call that answers when the user saves it, with its persona_id. Also get, request (a link for another account), sign_in (refresh the session) and record_login. A persona_id works on writ_scrape, writ_crawl_site, writ_browser_use and writ_run_workflow; a desktop persona ('device:...') signs in on that desktop, from its own vault.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "action": "<action>"
    },
    "name": "writ_personas"
  }
}

Input Schema

NameRequiredDescriptionDefault
whyNorequest: one line the user sees in the window.
waitNo
forceNo
actionYeslist (default) | get | request | wait | sign_in | record_login.
deviceNo
domainNolist: personas usable on this site. request: the site.
link_idNowait: the link_id of a persona_needed / request answer.
login_urlNo
persona_idNo
include_runsNo
execution_targetNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed6 schema fields changed
    • addedInput schema / properties / device
      Added value: +{
      +  "type": "string"
      +}
    • addedInput schema / properties / execution_target
      Added value: +{
      +  "enum": [
      +    "cloud"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / force
      Added value: +{
      +  "type": "boolean"
      +}
    • addedInput schema / properties / include_runs
      Added value: +{
      +  "type": "boolean"
      +}
    • addedInput schema / properties / login_url
      Added value: +{
      +  "type": "string"
      +}
    • addedInput schema / properties / wait
      Added value: +{
      +  "type": "boolean"
      +}
  2. Changed11 schema fields changed
    • changedInput schema / properties / action / description
      Previous value: -"What to do (default list). request = mint the persona link for a site (domain); wait = hold until the user saves it (link_id)."New value: +"list (default) | get | request | wait | sign_in | record_login."
    • removedInput schema / properties / device
      Removed value: -{
      -  "description": "A linked Writ desktop's agent_id (writ_devices) to act on. Omitted: the desktop this connection chose with writ_devices action='use', if any.",
      -  "type": "string"
      -}
    • changedInput schema / properties / domain / description
      Previous value: -"list: only personas usable on this host (suffix match), e.g. 'github.com'. With no match the answer is `persona_needed`, a request worded for the user. request: the site the new persona is for."New value: +"list: personas usable on this site. request: the site."
    • removedInput schema / properties / execution_target
      Removed value: -{
      -  "description": "list / request: 'cloud' asks for a Writ Cloud persona (cloud runs) even when this connection is scoped to a desktop.",
      -  "enum": [
      -    "cloud"
      -  ],
      -  "type": "string"
      -}
    • removedInput schema / properties / force
      Removed value: -{
      -  "description": "sign_in: re-run the login even when the current session still looks usable.",
      -  "type": "boolean"
      -}
    • removedInput schema / properties / include_runs
      Removed value: -{
      -  "description": "get: include the persona's recent runs (which workflows acted as it, and whether they succeeded).",
      -  "type": "boolean"
      -}
    • changedInput schema / properties / link_id / description
      Previous value: -"wait: the `link_id` of a persona_needed / request answer. Holds up to 75s and answers the moment the user saves the persona (persona_id), declines, or closes the window, else `still_waiting` (repeatable)."New value: +"wait: the link_id of a persona_needed / request answer."
    • removedInput schema / properties / login_url
      Removed value: -{
      -  "description": "record_login: exact sign-in page URL when known; defaults to the persona's domain root (the AI finds the form from there). list / request: the site's sign-in page, carried into `create_url` so the new persona can record it.",
      -  "type": "string"
      -}
    • removedInput schema / properties / persona_id / description
      Removed value: -"Which persona — required for get / sign_in / record_login."
    • removedInput schema / properties / wait
      Removed value: -{
      -  "description": "list + domain: one call holds up to 75s for a persona for that site to appear. A link_id is awaited with action='wait'.",
      -  "type": "boolean"
      -}
    • changedInput schema / properties / why / description
      Previous value: -"request: one line the user sees in the window: why the account is needed."New value: +"request: one line the user sees in the window."
  3. Changed2 schema fields changed
    • addedInput schema / properties / device
      Added value: +{
      +  "description": "A linked Writ desktop's agent_id (writ_devices) to act on. Omitted: the desktop this connection chose with writ_devices action='use', if any.",
      +  "type": "string"
      +}
    • addedInput schema / properties / execution_target
      Added value: +{
      +  "description": "list / request: 'cloud' asks for a Writ Cloud persona (cloud runs) even when this connection is scoped to a desktop.",
      +  "enum": [
      +    "cloud"
      +  ],
      +  "type": "string"
      +}
  4. First observed

TDQS

A4/5.0
Behavior5/5

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

Beyond annotations (readOnlyHint=false, destructiveHint=false, openWorldHint=true), the description discloses non-obvious behavior an agent could not infer: no password ever passes through the tool or the conversation, answers never prompt for credentials because they are typed only into a Writ window. It also reveals that action='wait' is a single held/blocking call that resolves when the user saves the link, and that a 'device:...' persona signs in on that desktop from its own vault. These are exactly the operational traits annotations cannot carry.

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?

The description is front-loaded with the concept and the credential guarantee before the action mechanics, and nearly every clause carries information — no filler sentences. It is dense as one unbroken paragraph with semicolon-chained action semantics, which makes scanning harder than a short action list would, but it stays proportionate to a six-action tool.

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

Completeness3/5

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

For a six-action, eleven-parameter tool with no output schema, the description covers the primary discovery flow (list → persona_needed → wait) and the cross-tool handoff thoroughly. It is silent on what get, sign_in and record_login actually return or change, and on the device, force and execution_target parameters, leaving real gaps for a tool whose actions have distinct outcomes.

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 description coverage is only 36% across 11 parameters, so the description needs to compensate, and it does for domain (list vs request), link_id (wait), persona_id (and which sibling tools accept it) and the device-prefix convention. However, force, login_url, include_runs, execution_target and device are left entirely undefined, and the schema's separate boolean 'wait' parameter versus action='wait' is never disambiguated — a real ambiguity for an agent. Meaning is added, but not enough for this coverage gap.

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

Purpose4/5

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

The description opens with a concrete framing — "The user's own accounts on websites (personas): how Writ signs in as them" — and then enumerates the six actions (list, get, request, wait, sign_in, record_login). It also positions persona_id as a downstream artifact for writ_scrape, writ_crawl_site, writ_browser_use and writ_run_workflow, so an agent can see where this tool sits. It stops short of a single crisp verb+resource statement, but the concept and action surface are unambiguous.

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

Usage Guidelines4/5

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

Usage is spelled out for the main path: action='list' with domain shows usable personas, and the empty case is described precisely (persona_needed → tell_user, create_url, link_id), followed by when to use action='wait'. That is real when-to-use guidance rather than an implied suggestion. What is missing is any differentiation among the remaining actions — get, sign_in and record_login get only parenthetical stubs, and no when-not-to-use condition is stated.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.