Skip to main content
Glama

Lovie Company Formation

Ingest Chat Evidence

fundraising_ingest_chat_evidence

Turn one message from a chat thread into a touch on the investor it is mapped to. Name the message by its id — you do not send its text: the server reads the thread from the source that holds it, under that source's own grant for your agent_name, so what is recorded is what the provider says was said. Refused on a thread the founder stopped, for an agent whose tier on it is none, and for a citation the thread does not hold. At the metadata tier no message text is stored at all; at the body tier the record keeps the cited message and only the turns immediately around it, never the thread's unrelated history. Every stored quote is screened for instructions aimed at an agent, and a verdict that could not be obtained is recorded as unscreened rather than as clean. The reply never carries the stored words back.

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": {
      "companyId": "<companyId>",
      "threadId": "<threadId>"
    },
    "name": "fundraising_ingest_chat_evidence"
  }
}

Input Schema

NameRequiredDescriptionDefault
kindNo
summaryNo
threadIdYesUUID value wrapper.
agentNameNo
companyIdYesUUID value wrapper.
citedExternalIdNo

Output Schema

NameRequiredDescriptionDefault
touchNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added

TDQS

A4.1/5.0
Behavior5/5

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

Annotations only cover openWorldHint=false and destructiveHint=false, and the description adds substantial context beyond them: the server reads the thread under the source's grant for agent_name, refusal cases, metadata-tier vs body-tier storage rules, screening for agent-directed instructions, recording unverifiable verdicts as 'unscreened', and that the reply never echoes stored words. This is rich behavioral disclosure that annotations alone don't provide.

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 core action is front-loaded in the first sentence and each subsequent sentence carries real information (input protocol, refusals, tier behavior, screening). It is dense but nearly every clause earns its place; a slightly tighter grouping of the tier/screening rules would push it higher.

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?

Output schema exists so return values needn't be described, and the description covers the non-obvious risks an agent needs: refusal conditions, storage-tier differences, prompt-injection screening, and the no-echo reply contract. Complete for a six-parameter, nested-object tool.

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?

At 33% schema coverage the description must compensate, and it does: threadId is explained as the message-id citation, agentName as the identity whose grant authorizes the read, citedExternalId as a citation the thread must hold, and companyId as the investor mapping. The 'kind' enum and summary field aren't clarified, so it's strong but incomplete.

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 specific verb+resource: 'Turn one message from a chat thread into a touch on the investor it is mapped to,' which distinguishes it from generic touch-recording siblings by naming the message-id input and server-side thread read. It doesn't explicitly contrast with close siblings like fundraising_record_touch or fundraising_capture_what_was_said, but the scope is concrete enough to select correctly.

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?

It states refusal conditions (founder-stopped thread, agent tier 'none', citation not in thread), which effectively communicate when the tool won't work, but it never routes the agent to an alternative when those conditions fail nor contrasts with sibling ingestion tools. Usage is implied rather than explicitly guided.

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.