Sasha, Neural Partners Chief of Staff

Neural Labs pricing discovery

pricing_discovery

Sasha discusses what a Neural Labs agency or advisory engagement would cost. She asks about scope before naming a figure and quotes only Neural Partners' approved starting points, each of which depends on the work. structuredContent.prices carries exactly the approved figures her reply quoted, from configuration, with their conditions. Sasha, Neural Partners' AI agent on Claude, can be wrong. Prices are starting points only. Chats are kept up to 90 days. She never books or submits anything.

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": {
      "message": "<message>"
    },
    "name": "pricing_discovery"
  }
}

Input Schema

Name	Required	Description	Default
`message`	Yes	The pricing question, or what the work is, in plain text.
`conversation_id`	No	The conversation_id an earlier call returned. Omitted, a new conversation starts. Earlier turns are kept server-side, so only the new message is sent.

Output Schema

Name	Required	Description
`error`	No	Present only on a result with isError: true.
`links`	No	The Neural Partners pages Sasha's reply pointed at, as absolute URLs. Only pages on the Neural Partners site are ever listed. Empty when she pointed at none.
`reply`	No	Sasha's reply, on its own. The disclosure is never part of it.
`prices`	No	The approved Neural Partners starting prices Sasha's reply quoted, read from Neural Partners' configuration rather than from the reply text. Empty when she quoted none.
`degraded`	No	True when the reply is a stand-in because Sasha ran out of time; the same message can be sent again.
`disclosure`	No	Present on the first reply of a conversation: what the person should know about Sasha. Also sent as its own first content block.
`offer_handoff`	No	True on the reply where Sasha offers to pass the conversation to the Neural Partners team.
`conversation_id`	Yes	Continues this conversation on a later call.

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 22, 2026

TDQS

A4/5.0

Behavior5/5

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

Beyond the annotations, the description discloses the call's internal behavior (asks about scope before quoting a figure), the provenance of quoted figures (Neural Partners' approved starting points, sourced from configuration), the output contract (structuredContent.prices carries exactly the quoted figures with their conditions), a reliability caveat, a 90-day chat retention policy, and the hard boundary that it never books or submits. This is far more behavioral context than readOnlyHint/destructiveHint/openWorldHint convey; readOnlyHint=false is not contradicted, since chats are persisted for 90 days.

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?

Purpose and scope-first behavior are front-loaded in the opening sentence, and each later sentence adds a distinct fact (provenance, caveat, retention, no booking). The third-person 'Sasha' narration and the agent-persona framing make it slightly more discursive than necessary, but there is no true filler.

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?

With an output schema present, the description only needs to signal what structuredContent.prices means, which it does, and parameter semantics are fully covered by the schema. Retention, reliability, and the no-booking boundary round out everything an agent needs to call this correctly and to describe results to a user.

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 100% for both parameters, so the schema already explains 'message' and the conversation_id resume semantics, including that prior turns are held server-side. The description adds nothing about either parameter — notably it never mentions conversation continuity, so it does not earn above the baseline.

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 states a specific activity — discussing Neural Labs agency/advisory engagement cost, scoping first, quoting only approved starting points — so an agent knows this is a pricing-quote conversation tool, not a general chat. It implicitly separates itself from book_discovery with 'She never books or submits anything,' but never distinguishes itself from the near-identical sibling ask_sasha, which is the ambiguity an agent would most likely hit.

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?

Usage is only implied: 'the pricing question, or what the work is' in the schema plus the pricing focus of the description suggest when to call it. There is no explicit statement like 'for non-pricing questions use ask_sasha,' and no exclusions or prerequisites are named. With a sibling as close as ask_sasha, the routing guidance is left to inference.

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

All Sasha, Neural Partners Chief of Staff 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.

Resources

Need Help?