Skip to main content
Glama

Track Prospects

track_prospects

A person belongs to exactly one campaign, so someone already tracked in another campaign is reported in skipped rather than added twice; pass move_existing=True to move them into this agent instead. A durable contact left without a campaign (its campaign was deleted) is instead adopted straight in and reported in adopted. Someone already on THIS campaign whom the user stopped pursuing (a 'skipped' stage) is reported in removed, not re-enrolled and not sent to — clear the skip with update_prospect first if the user explicitly wants them back.

On an agent with a define_sequence flow, tracking IS enrollment — the flow fires each prospect's first touch itself (see the returned flow field); never queue a first send for prospects you just tracked. Dict with created and updated counts, and moved, adopted, removed and skipped details

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
itemsYesPeople to add (see ProspectItem). Each carries a `person` with the identifiers you already hold (LinkedIn URL / slug, email, or provider_id) copied verbatim from the tool result that surfaced them — never a slug rebuilt from a display name, which resolves to the wrong person. Junk like a company URL or 'N/A' is rejected. The display name goes on `person.display_name`, taken from the same row as the identifier — a name that shares no token with the LinkedIn name already on file for its ID rejects the whole call. Optional freeform `data` per item.
agent_idYesID of the agent to add prospects to
move_existingNoMove people tracked in another campaign into this one instead of skipping them. Destructive — it cancels whatever was still queued for them, so only pass it on an explicit user request.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / items / description
      Previous value: -"People to add (see ProspectItem). Each carries a `person` with the\nidentifiers you already hold (LinkedIn URL / slug, email, or\nprovider_id) copied verbatim from the tool result that surfaced them —\nnever a slug rebuilt from a display name, which resolves to the wrong\nperson. Junk like a company URL or 'N/A' is rejected. The display name\ngoes on `person.display_name`; optional freeform `data` per item."New value: +"People to add (see ProspectItem). Each carries a `person` with the\nidentifiers you already hold (LinkedIn URL / slug, email, or\nprovider_id) copied verbatim from the tool result that surfaced them —\nnever a slug rebuilt from a display name, which resolves to the wrong\nperson. Junk like a company URL or 'N/A' is rejected. The display name\ngoes on `person.display_name`, taken from the same row as the identifier\n— a name that shares no token with the LinkedIn name already on file for\nits ID rejects the whole call. Optional freeform `data` per item."
  2. Changed3 schema fields changed
    • addedInput schema / properties / agent_id
      Added value: +{
      +  "description": "ID of the agent to add prospects to",
      +  "type": "integer"
      +}
    • removedInput schema / properties / task_id
      Removed value: -{
      -  "description": "ID of the task to add prospects to",
      -  "type": "integer"
      -}
    • changedInput schema / required
      Previous value: -[
      -  "task_id",
      -  "items"
      -]New value: +[
      +  "agent_id",
      +  "items"
      +]
  3. First observed

TDQS

A5/5.0
Behavior5/5

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

Annotations are only readOnlyHint=false and destructiveHint=false, which are generic. The description compensates thoroughly: it discloses that move_existing is destructive, explains the skipped/adopted/removed semantics, and clarifies the define_sequence enrollment behavior. It also notes the returned flow field. No contradiction with annotations; instead it adds rich behavioral context.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Although long, every sentence earns its place. The summary is front-loaded with the core purpose, followed by structured paragraphs covering dedup behavior, edge cases, and the sequence flow. No fluff; the density is justified by the complexity of the tool. It is well-organized for an agent to parse.

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?

The description covers all important scenarios: cross-agent dedup, within-campaign skip stages, adopted orphans, removed entries, the destructive move_existing, and the define_sequence enrollment nuance. It also provides a returns tag summarizing counts for created/updated/moved/adopted/removed/skipped. With no output schema, this high-level return description is sufficient. Nothing critical is missing.

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

Parameters5/5

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

Schema description coverage is 100%, but the description adds significant semantics beyond the schema: it details the person identifier requirements (copy verbatim, never rebuild slug, reject company URL/N/A), the data field signal usage ('Hiring for Founding AE'), and move_existing's destructive effect. This goes well beyond what the schema provides, making parameter usage crystal clear.

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?

The description opens with a clear, specific statement: 'Bulk-add people (prospects) to an agent's outreach list with cross-agent dedup.' This names the exact verb, resource, and a key feature (dedup). It also distinguishes itself from sibling track_monitored_companies by explicitly directing to that tool for companies, so there is no ambiguity.

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?

The description gives direct usage guidance: 'For companies to monitor, call track_monitored_companies instead.' It explains when to pass move_existing ('on an explicit user request'), the behavior for skipped/adopted/removed prospects, and the important rule about the define_sequence flow (tracking IS enrollment, never queue a first send). This is explicit and actionable, leaving no inference to the agent.

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.

Resources