Reach MCP

create_job

Queue a batch of LinkedIn actions (messages, invitations, profile visits or comments) that Reach executes on its own over the following hours or days: spread across the account's activity window (set per account with update_account_quotas; default 09:00-18:00, Monday to Friday, in the account's timezone; overridable per job) with human-like gaps, never beyond the daily quota of the action. Returns immediately with a job_id and the planned schedule; the agent does not have to stay alive. The job pauses by itself when a quota is reached or the account disconnects and resumes when it can; webhooks job.started / job.progress / job.paused / job.completed report what happens. Use it for anything above a handful of actions instead of calling send_message or send_invitation in a loop. Always show the user the items before queuing them. Use for any batch above a handful of actions; returns at once with the schedule. Show the items to the user first. Not for a single action, call the direct tool.

Input Schema

TableJSON Schema

Name	Required	Description
`days`	No	Weekdays the job may run: mon, tue, wed, thu, fri, sat, sun. Default: the account's window (Monday to Friday unless changed).
`items`	Yes	One object per action, in the order to execute them; fields depend on the action (see action).
`label`	No	Short free text shown in the dashboard and in webhook events, e.g. 'Follow-ups week 40'.
`action`	Yes	What every item does. send_message: text + conversation_linkedin_id (reply) or recipient_linkedin_id (new thread). send_invitation: linkedin_id_or_url + optional message (<200 chars). visit_profile: linkedin_id_or_url. comment_post: post_url_or_urn + text.
`timezone`	No	IANA zone such as Europe/Paris. Default: the account's window setting, else the zone of its proxy country.
`account_id`	Yes	Reach id of the LinkedIn account to act on, from list_accounts.
`window_end`	No	HH:MM in the job's timezone; default: the account's window (18:00 unless changed).
`window_start`	No	HH:MM in the job's timezone; default: the account's window (09:00 unless changed).
`idempotency_key`	No	Optional. A key you choose (a UUID is fine) that names this exact call. If you retry with the same key and the same arguments, the first call's result is returned and nothing is done twice on LinkedIn. Reusing a key with different arguments is refused. Keys expire after 24 hours.

Output Schema

TableJSON Schema

Name	Required	Description
`id`	No
`items`	No
`label`	No
`action`	No
`source`	No
`status`	No
`totals`	No
`window`	No
`resume_at`	No
`account_id`	No
`created_at`	No
`started_at`	No
`completed_at`	No
`next_item_at`	No
`pause_reason`	No	quota_reached \| account_disconnected \| billing \| manual

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedSep 27, 2026

changedInput schema / properties / action / description

Previous value: -"What every item does. send_message: text + conversation_linkedin_id (reply) or recipient_linkedin_id (new thread). connect: linkedin_id_or_url + optional message (<200 chars). visit_profile: linkedin_id_or_url. comment_post: post_url_or_urn + text."New value: +"What every item does. send_message: text + conversation_linkedin_id (reply) or recipient_linkedin_id (new thread). send_invitation: linkedin_id_or_url + optional message (<200 chars). visit_profile: linkedin_id_or_url. comment_post: post_url_or_urn + text."

changedInput schema / properties / action / enum

Previous value: -[
-  "send_message",
-  "connect",
-  "visit_profile",
-  "comment_post"
-]New value: +[
+  "send_message",
+  "send_invitation",
+  "visit_profile",
+  "comment_post"
+]

AddedSep 27, 2026

TDQS

A4.7/5.0

Behavior5/5

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

Annotations only declare the safety profile; the description goes far beyond, disclosing asynchronous execution, spreading across the account activity window with human-like gaps, quota-bounded behavior, self-pausing on quota/disconnect and self-resuming, webhook lifecycle events (job.started/progress/paused/completed), and that it returns immediately with a job_id and schedule so the agent need not stay alive.

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

Conciseness3/5

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

The core is well front-loaded, but the closing sentences ('Use for any batch above a handful of actions; returns at once with the schedule. Show the items to the user first. Not for a single action, call the direct tool.') restate content already delivered earlier in the paragraph, adding length without new information.

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?

Given an output schema exists, the description does not need to detail return values, and it still notes the job_id/schedule return and webhook reporting. Async lifecycle, quota interaction, defaults and prerequisites are all covered for a 9-parameter batching tool.

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 already 100%, so the baseline is 3, but the description adds meaning the schema does not: it ties window_start/window_end/timezone/days to the account-level setting managed by update_account_quotas, explains the default fallback chain, and clarifies the action-to-item-fields relationship ('fields depend on the action').

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 and resource ('Queue a batch of LinkedIn actions ... that Reach executes on its own over the following hours or days') and enumerates the action types. An agent can distinguish this asynchronously-executed batching tool from the synchronous send_message/send_invitation siblings without opening any 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?

Explicitly names the alternatives ('instead of calling send_message or send_invitation in a loop'), gives a threshold ('anything above a handful of actions'), an exclusion ('Not for a single action, call the direct tool') and a workflow rule ('Always show the user the items before queuing them'). Nothing is left to inference.

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

All Reach MCP 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?