Skip to main content
Glama

Request an intro

requestIntro
DestructiveIdempotent

Ask Clera to introduce a candidate to your company for one of your open roles. Can start candidate outreach and submit an application to a connected hiring system. Target either a candidate from searchCandidates (talentId, with jobId or the interactive role picker) or an entry from listPendingIntroRequests / listRecommendedCandidates (its opportunityId, with talentName and roleName copied verbatim from that entry). When the user has not named a role, call this WITHOUT jobId and they pick the role in an interactive form; submitting it sends the request. reason carries only what the user said; your own view of the candidate goes in assistantNote. Fit flags are never a gate: a conflict flag is something to tell the user, not a reason to withhold the call. The response carries the flags as they stand at request time; pass them on. Re-requesting the same candidate for the same role sends nothing new and is safe. Refer to candidates and roles by name; the user never sees ids.

Clera privately records this tool's inputs and results for usage tracking and support.

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": "requestIntro"
  }
}

Input Schema

NameRequiredDescriptionDefault
jobIdNoThe open role to request the intro for. Resolve the id yourself with the job tool's list command when the user already named a role. When the role is not settled, omit it and CALL THE TOOL ANYWAY; the user gets an interactive role picker. Never list roles in chat and never ask the user for an id.
orgIdNoOrganization to act as. Only needed when you belong to more than one organization.
reasonNoWhat the user said about why they want this candidate, in their own words, quoted or lightly trimmed and in their voice, e.g. 'The HPC master's is interesting for our infra work.' No lead-in such as 'the hiring manager said' or 'per the user'. Never fill it from the profile, fit flags, fit headline or your own reading of the candidate; that belongs in assistantNote. When the user gave no reason, ask once if the conversation allows it, otherwise leave it out. Stored on the opportunity as the company's reason.
categoryNoOnly when the user's own words name it: strong_stack_match (tech stack fits) | great_trajectory (career growth) | domain_fit (industry experience aligns) | founder_energy (startup mindset). Leave it out otherwise.
roleNameNoWhen acting on a listed opportunity, the role name copied verbatim from the same row. Checked against the record; a mismatch is refused.
talentIdNoThe candidate's id, exactly as returned by searchCandidates. Pair it with jobId.
talentNameNoWhen acting on a listed opportunity, the talent's name copied verbatim from that row. Checked against the record; a mismatch is refused.
assistantNoteNoYour own remarks for Clera, kept apart from the user's reason and never shown as the company's view: what the profile or fit flags say, a profile fact that contradicts what the user said, something to check. Leave it out when you have nothing to add.
opportunityIdNoAlternative target: the opportunity id from a listPendingIntroRequests or listRecommendedCandidates row.

Output Schema

NameRequiredDescriptionDefault
flagsYesFit flags for this candidate against the role, computed at request time.
messageYes
outcomeYes
roleNameYes
opportunityYes

Schema Changelog

Changes observed during successful MCP inspections.

No schema history has been recorded yet.

TDQS

A4.4/5.0
Behavior3/5

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

Annotations already declare readOnly=false, destructive=true, idempotent=true, openWorld=true, and the description reinforces the idempotency claim ('re-requesting the same candidate for the same role sends nothing new and is safe') plus discloses that inputs/results are recorded and that mismatched names are refused. However, it never explains the more consequential side of destructive/openWorld — what outreach actually gets sent, to whom, or whether an erroneous request can be retracted — which is exactly where an agent needs behavioral detail. Solid but not full credit against a destructive, open-world write.

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 purpose, then target selection, then working rules; nearly every sentence carries an instruction an agent would otherwise get wrong. It is dense and long for a nine-parameter tool, and the fit-flag sentences could be tightened, but there is no 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?

Nine parameters, zero required, an output schema, and multiple targeting paths — the description covers all of it: both entry modes, the interactive picker fallback, the flag pass-through on the response, refusal on name mismatch, and idempotent retries. An agent has enough to call this correctly in every branch.

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?

Schema coverage is 100%, so the baseline is 3 and each field is already documented individually. The description still adds group-level semantics the schema lacks: which parameters pair together for each targeting mode, and the strict separation of user-voiced reason from the agent's own assistantNote. That cross-parameter guidance is genuine value beyond the per-field text.

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 concrete verb and resource: 'Ask Clera to introduce a candidate to your company for one of your open roles,' then immediately scopes the two side effects (candidate outreach, application submission). It names the sibling tools that supply each target (searchCandidates, listPendingIntroRequests, listRecommendedCandidates), so an agent can distinguish it from search/outreach siblings without opening a 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?

Explicit routing rules: use talentId+jobId for a searchCandidates candidate, opportunityId+verbatim talentName/roleName for a listed opportunity, and call without jobId to trigger the interactive picker. It also excludes usage ('Fit flags are never a gate'; 'Never list roles in chat and never ask the user for an id') and reassures on retries.

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