Clarifo

open_in_clarifo

Read-onlyIdempotent

Show Clarifo content in Clarifo's panel, beside the conversation. It suits a user working in Clarifo: asking about a company, their workspace documents, portfolios or notes, or reading an answer that cites filings, whose sources the panel shows next to the answer. It only shows: read data with get_workspace_item, get_filing_snapshot, portfolios and get_company_snapshot. Give at most one target:

sources: an answer's cited pages, in the answer's order, as [{"page_id": ..., "highlight": }] (up to 20), once per answer with every page_id it cites. The user reads the source and the answer side by side: the first page opens highlighted and they page through the rest. The numbered links in the answer stay.
item_id: a document from list_workspace_items. The panel shows it; the user can edit a paragraph there and click a citation [n] to see its page.
page_id (+ highlight, the cited figure or sentence): a filing page, as for get_filing_snapshot. The panel shows the page highlighted.
portfolio_id (+ period 1M, 3M, 6M, 1Y or YTD): one of the user's lists.
company (name or ticker): its profile: what it does, key figures by year and the latest quarter, price, profile sections and latest filings. Put the other companies of the conversation in companies (up to 8) so the user can switch between them.
note_id: one of the user's notes (from notes); the user can add to it, edit it or ask for a document from it there.
notes_company (name or ticker): the user's notes on that company, in the panel's Notes tab. Use it when the user asks what they noted about a company: answer in a sentence or two and open them here.
nothing: the panel's home (portfolios, documents, automations).

display: "inline" (a card in the conversation, the default) or "fullscreen" (the whole window), only when the user asks for one. The panel asks the host for it; the host decides, and the user can switch in the panel either way.

Say in one sentence what opened; a sources open needs no words. When the user says they saved an edit in the panel, do not apply it again.

Returns: {"open": "document" | "source" | "sources" | "portfolio" | "company" | "note" | "notes" | "portfolios", "document" | "source" | "portfolio" | "company" | "portfolios": <the data tool's result>}

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "open_in_clarifo"
  }
}

Input Schema

Name	Required	Default
`period`	No	3M
`company`	No
`display`	No
`item_id`	No
`note_id`	No
`page_id`	No
`sources`	No
`companies`	No
`highlight`	No
`portfolio_id`	No
`workspace_id`	No
`notes_company`	No

Output Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 9, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Annotations only declare generic safety (readOnly, idempotent, non-destructive), but the description adds real behavioral context beyond them: the panel delegates to the host ('The panel asks the host for it; the host decides'), the user can switch display mode regardless, sources open with the first page highlighted and numbered links preserved, and 'When the user says they saved an edit in the panel, do not apply it again.' These are operational rules an agent could not infer from annotations.

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?

Front-loaded with the one-line purpose and the read-only constraint, then organized as a scannable bullet list of targets. It is long, but the length is largely earned given twelve parameters and zero schema documentation. A few clauses are redundant with the Returns block and with each other (highlight is explained twice).

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?

For a twelve-parameter panel-opening tool, an agent has everything it needs: which target to pick for each user intent, payload formats and caps, default display behavior, a caution against double-applying edits, and a Returns block describing the open/discriminator shape. Output schema exists, so return values are covered rather than a gap.

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

Parameters4/5

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

With 0% schema description coverage the description carries the full burden, and it mostly does: it gives the sources payload shape and a 20-item cap, the period enum (1M/3M/6M/1Y/YTD), the companies cap of 8, the two display values and a default, and the identity and use of each target id. The one gap is workspace_id, which appears in the schema but is never mentioned, and the 'at most one target' rule is stated without saying what happens if several are passed.

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?

Opens with a specific verb and resource ('Show Clarifo content in Clarifo's panel, beside the conversation') and immediately contrasts with the data tools: 'It only shows: read data with get_workspace_item, get_filing_snapshot, portfolios and get_company_snapshot.' An agent can tell this is a presentation tool, not a retrieval tool, 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 Guidelines5/5

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

Gives explicit when-to-use conditions for each target type (sources for a cited answer, notes_company when the user asks what they noted, nothing for the panel home), states which sibling tools supply the data, and names when to use display='fullscreen' ('only when the user asks for one'). Guidance is not left implicit anywhere.

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

All Clarifo tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.