Sliq

Register Manual Linkedin Invites

register_manual_linkedin_invites

Only registers invites still pending; anything else comes back as not_pending — which can mean already accepted or never sent, so don't assume acceptance (check the prospect's linkedin_stage before messaging). Unmatched prospects that have a LinkedIn URL but no provider_id yet come back under resolving: Sliq looks up their profile in the background to identify them (in progress, not failed). Attach a resolved_profile trigger to act on each result — an already-connected one sends no acceptance event, so it won't engage on its own. Use this only after the prospects are tracked and the user has confirmed they sent the requests themselves.

Usually returns in seconds: the pending-invitations read stops as soon as every named prospect is matched, and recently hand-sent invites sit at the top of the list. When some prospects aren't found there, it pages the user's full backlog with paced gaps to protect their LinkedIn account — up to a couple of minutes on a large backlog — so warn the user about the possible wait only when registering invites sent long ago or likely already accepted, not for a just-sent batch.

Pass the specific people the user named in identifiers. Only set all_tracked=True when the user said they hand-invited the whole campaign — it registers every tracked prospect that has a real pending invitation, which can pull in an unrelated pending invite, so it's a deliberate opt-in rather than the default. Dict with registered / already_registered / not_pending / resolving / ambiguous / skipped lists, agent_has_accept_trigger (whether the agent has an accept trigger — a sequence campaign advances on accept via its flow regardless), and a summary message.

Input Schema

TableJSON Schema

Name	Required	Description
`node_id`	No	Optional. The sequence-DAG connection_request node the hand-sent CR satisfies (mirrors setup_linkedin_sequence's node_id) — stamped on the recorded row so its acceptance advances the flow onto the next node. Omit to auto-detect the agent's sole CR node; pass it explicitly on a multi-CR sequence. Rejected with ModelRetry if it isn't a connection_request node in the agent's sequence.
`agent_id`	Yes	The agent whose tracked prospects were manually invited.
`all_tracked`	No	Register every tracked prospect in the agent — use only when the user invited the whole campaign by hand.
`identifiers`	No	LinkedIn URLs, provider_ids, or names of the people the user invited by hand. Required unless `all_tracked` is set.
`on_behalf_of`	No	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.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 19, 2026

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."
+}

Changed5 schema fields changedSep 23, 2026

addedInput schema / properties / agent_id

Added value: +{
+  "description": "The agent whose tracked prospects were manually invited.",
+  "type": "integer"
+}

changedInput schema / properties / all_tracked / description

Previous value: -"Register every tracked prospect in the task — use only when the user\ninvited the whole campaign by hand."New value: +"Register every tracked prospect in the agent — use only when the user\ninvited the whole campaign by hand."

changedInput schema / properties / node_id / description

Previous value: -"Optional. The sequence-DAG connection_request node the hand-sent CR\nsatisfies (mirrors setup_linkedin_sequence's node_id) — stamped on the recorded\nrow so its acceptance advances the flow onto the next node. Omit to auto-detect\nthe task's sole CR node; pass it explicitly on a multi-CR sequence. Rejected with\nModelRetry if it isn't a connection_request node in the task's sequence."New value: +"Optional. The sequence-DAG connection_request node the hand-sent CR\nsatisfies (mirrors setup_linkedin_sequence's node_id) — stamped on the recorded\nrow so its acceptance advances the flow onto the next node. Omit to auto-detect\nthe agent's sole CR node; pass it explicitly on a multi-CR sequence. Rejected with\nModelRetry if it isn't a connection_request node in the agent's sequence."

removedInput schema / properties / task_id

Removed value: -{
-  "description": "The task whose tracked prospects were manually invited.",
-  "type": "integer"
-}

changedInput schema / required

Previous value: -[
-  "task_id"
-]New value: +[
+  "agent_id"
+]

First observedSep 15, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Annotations only declare readOnlyHint=false and destructiveHint=false; the description goes far beyond by disclosing that it cancels the now-redundant queued connection request, that unmatched prospects return under `resolving` as in-progress background lookups, that `not_pending` is ambiguous, and the timing behavior (seconds normally, up to a couple of minutes when paging a large backlog, with paced gaps to protect the account). This is unusually rich context for a mutation tool.

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 one-line summary before the operational detail, and the returns block is clearly separated. It is long, but the length tracks the genuine complexity; a few clauses (e.g. the repeated restatement of the accept-trigger caveat) could be tightened.

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?

Covers the return lists, the `resolving` in-progress state, trigger attachment advice, timing expectations, and the all_tracked caveat. Given there is no output schema, the description supplies everything an agent needs to interpret results and call the tool correctly.

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 semantic value for `identifiers` (pass the specific people the user named) and `all_tracked` (deliberate opt-in with the risk of an unrelated pending invite). It adds little for `node_id` and `on_behalf_of`, which the description does not touch, so it stops short of a 5.

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: registers manually-sent LinkedIn connection requests so Sliq runs the accept→message sequence. It clearly distinguishes itself from siblings like manage_linkedin_invite_queue and setup_linkedin_sequence by describing reconciliation of hand-sent invites against real pending invitations.

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?

Gives explicit preconditions ('use this only after the prospects are tracked and the user has confirmed they sent the requests themselves') and explicit selection rules between `identifiers` and `all_tracked`, including the downside of the latter (can pull in an unrelated pending invite). It also tells the agent not to assume acceptance and to check linkedin_stage first.

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

All Sliq 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?