Skip to main content
Glama

Sliq

by com.getsliq
OverviewDiscussionsAdmin

Find Warm Intro Paths

find_warm_intro_paths
Read-only

Pass targets as a list of LinkedIn identifiers — profile URLs, username slugs, or provider_ids — for the people the user wants to reach. Runs in the background, pacing 1-5 minutes between targets to stay within LinkedIn's safety limits; results land as one row per target under list_name in the agent's entity store — read them with query_search_results(agent_id, where_clause="list_name = '<list_name>'"). Each row carries the mutual connections — the shared 1st-degree connections who can introduce the user.

For each target the search returns at most 10 mutual connections. A target that comes back with all 10 may have more the search didn't surface — treat that target's list as a sample rather than the complete set, and say so when you report it.

list_name is the bucket the target rows land in. Pass a short descriptive slug naming this set of targets (e.g. 'acme-intros', 'series-a-leads'); distinct slugs let one agent hold several independent warm-intro searches, and reusing a slug accumulates into one list.

Each target costs 1 profile lookup + 1 LinkedIn search against the daily budgets (~50-60 lookups for an established account, a quarter of that while a new one ramps up; the LinkedIn search budget is shared with all people searches), plus 1 credit only when a mutual connection is found for that target. An already-connected target with shared connections still costs that credit — the warmer path is still worth surfacing. Checking a target also links it to its primary employer's company, which may need a one-time web-domain lookup: 0.05 credits the first time a given company is looked up (cached after, so it never charges twice for the same company). Targets beyond today's budget are deferred — re-run tomorrow to continue. Confirm with the user before passing a large list. Only one warm-intro search can run at a time per account. A status dict in one of three forms. {'success': True, 'status': 'running', ...} on kickoff {'success': True, 'status': 'queued', ...} when another warm-intro search already holds the account's single slot — this one is queued behind it and starts automatically when the active one finishes {'success': False, 'error': str} on validation / budget failure

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
targetsYesLinkedIn identifiers (profile URLs, slugs, or provider_ids) of the people to find warm-intro paths to.
list_nameYesshort kebab slug naming this set of targets. Required.
on_behalf_ofNoAn agency client's email, from list_agency_clients, to run this call in that client's account: their agents, settings, connected accounts, and credits. Omit it to act on your own account.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / on_behalf_of
      Added value: +{
+  "anyOf": [
+    {
+      "type": "string"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "An agency client's email, from list_agency_clients, to run this call in that client's account: their agents, settings, connected accounts, and credits. Omit it to act on your own account."
+}
  2. Changed1 schema field changed
    • changedInput schema / properties / list_name / description
      Previous value: -"short kebab slug naming this set of targets; distinct slugs\nrender as separate Output sub-pills. Required."New value: +"short kebab slug naming this set of targets. Required."
  3. First observed

TDQS

A4.6/5.0
Behavior5/5

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

Extremely rich: discloses background execution, 1-5 minute pacing to respect LinkedIn safety limits, per-target credit costs, daily budget consumption and deferral, the one-at-a-time account slot with queuing, and the 10-mutual-connection sampling caveat. This goes well beyond the lone readOnlyHint annotation.

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 a summary, then organized by topic (targets, background/read-back, sampling caveat, slug semantics, costs). It is long but nearly every sentence carries operational weight; only minor trimming is possible.

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 complex async, credit-consuming, budget-limited tool with no formal output schema, the description covers return states (running/queued/error via the returns block), costs, deferral, and result retrieval. Nothing an agent needs to invoke it correctly appears missing.

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, but the description adds real meaning: it clarifies target identifier formats, explains that list_name is a bucket slug whose reuse accumulates results, and points to list_agency_clients for on_behalf_of. It adds value beyond the schema 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?

States a specific verb + resource + scope: find which of the user's 1st-degree LinkedIn connections can introduce them to each target person. This is clearly distinguishable from siblings like search_linkedin_connections and search_linkedin_people, and the opening summary line makes the goal unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

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

Explains what targets to pass, that it runs in the background with pacing, and directs the agent to read results via query_search_results. It advises confirming with the user before a large list. However, it never explicitly contrasts this with alternative people/connection search tools, so the when-not guidance is incomplete.

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

All Sliq toolsInspect
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