Skip to main content
Glama

create_signal

Create a signal: the account checks one thing on LinkedIn on a schedule and records every new item it finds as an event, which you read with get_signal_events. The first check records a silent baseline: what is already there is not reported, except that personPosts, companyPosts and postSearch signals report posts published since the signal was created. Creating a signal that already exists on the account (same type and target) returns that signal instead of a second one; a paused one stays paused. Pass only the target fields of the chosen type:

  • network: no target. Reports connectionAccepted, connectionAdded and connectionRequestReceived. Checked every hour; spends nothing from account limits.

  • profileViewers: no target. Reports profileViewed. Checked every 3 hours; spends nothing from account limits.

  • personPosts: personUrl. companyPosts: companyUrl. Reports postPublished. Checked once or twice a day. The first check opens the profile or company page once (1 of stPersonProfileViews or stCompanyPageViews, and the person may see the account among their profile viewers); later checks mostly run as LinkedIn searches and spend some stSearchQueries – usually a few a day for the whole account – plus occasional visits.

  • postSearch: term (1 to 3 phrases) and an optional filter. Reports postFound for new posts matching any phrase. Checked once or twice a day; each check spends 1 of stSearchQueries. By default an account has 300 searches a month and each postSearch signal spends 30 to 60 of them, so put the wordings of one topic into one signal rather than a signal per phrase.

  • postEngagement: postUrl, or myPosts: true for the posts the account publishes. Reports commentAdded and reactionAdded. Spends nothing from account limits, apart from occasional visits. A postUrl signal watches the post for 30 days, then stops. Fails with signalLimitReached when the account already has the maximum number of active signals of this type: pause or delete one first.

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": {
      "type": "<type>"
    },
    "name": "create_signal"
  }
}

Input Schema

NameRequiredDescriptionDefault
termNopostSearch only, required there. 1 to 3 phrases, each up to 50 characters and with no quotation marks inside; a single phrase is a one-item array. Each phrase is searched as an exact phrase, and a post containing any of them is reported.
typeYesWhat to watch: network, profileViewers, personPosts, companyPosts, postSearch or postEngagement.
labelNoOptional. Your own identifier, up to 64 characters, such as a CRM record ID. It comes back with the signal and with every event it produces.
filterNopostSearch only, optional. Narrows the search; a post must match every field you set. The signal always looks for the newest posts, so there is no sort or date filter.
myPostsNopostEngagement only. true to watch comments and reactions on the account's own posts. Pass postUrl or myPosts, not both.
postUrlNopostEngagement only. URL of the post whose new comments and reactions to report. Pass postUrl or myPosts, not both.
personUrlNopersonPosts only, required there. LinkedIn profile URL of the person (e.g., 'https://www.linkedin.com/in/john-doe').
companyUrlNocompanyPosts only, required there. LinkedIn company page URL (e.g., 'https://www.linkedin.com/company/example').

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added

TDQS

A4.9/5.0
Behavior5/5

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

Annotations only declare readOnlyHint=false and openWorldHint=true; the description goes far beyond that. It discloses the silent first-check baseline, the exception for personPosts/companyPosts/postSearch, per-type check frequency, quota spend (stSearchQueries, stPersonProfileViews, stCompanyPageViews), the 300-searches/month default and 30-60 search cost per postSearch signal, and the 30-day expiry of postUrl signals.

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 opening sentence is somewhat long and nested, but the body is tightly organized as one bullet per signal type, each carrying a distinct fact (target, events, frequency, cost). Given six modes plus quota and dedup behavior, the length is largely earned, though the density could be trimmed slightly.

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 mutation tool with eight parameters, nested filter objects, and no output schema, the description covers everything an agent needs: what gets created, what the first check does, what events flow to get_signal_events, what each type costs, dedup semantics, and the limit failure mode.

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 coverage is already 100%, but the description adds conditional meaning the schema cannot express: which target fields apply to which type, that filter and term are postSearch-only, and the operational advice to merge one topic's phrases into a single signal rather than one signal per phrase to conserve search quota. That is genuine guidance beyond field-level docs.

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 first sentence states a specific verb and resource ('Create a signal') and immediately defines what a signal is: a scheduled LinkedIn check that records new items as events read via get_signal_events. The six signal types are then enumerated with their targets, so an agent can distinguish this from get_signal, pause_signal, and the rest of the signal family without opening the 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?

Usage is spelled out per type: which target fields to pass for each of the six types ('Pass only the target fields of the chosen type'), when a target is absent (network, profileViewers), and the postUrl-vs-myPosts choice. It also states the dedup rule (same type and target returns the existing signal, a paused one stays paused) and the recovery path when signalLimitReached fires: pause or delete one first.

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.