MCP Tools: Automation

Technische Referenz auf Englisch.

Agents

Create/manage AI agents, agent memory and learnings, agent chains (beta)

list_agents

read-only

List all agents in this workspace with their status, assigned tools, and skill count. Use get_agent for full details of a specific agent.

Parameter	Type	Required	Description
enabled_only	boolean	no

get_agent

read-only

Get full details of a subagent including system prompt and configuration.

Parameter	Type	Required	Description
slug	string	yes	Agent slug (e.g. sales-researcher)

create_agent

action

Create a new subagent for this workspace. The agent gets its own system prompt, model, and tool access controls.

Example: Create a ‘Lead Qualifier’ agent that only has access to research + qualification tools, with a specialized prompt for evaluating company fit.

Trigger types:

• keyword: { type: ‘keyword’, pattern: ‘qualify *’ } — invoked when message matches pattern
• event: { type: ‘event’, event: ‘lead_replied’ } — invoked on system event
• schedule: { type: ‘schedule’, cron: ‘0 9 * * 1’ } — runs on schedule
• channel: { type: ‘channel’, channel_id: ‘uuid’ } — default agent for a channel

Parameter	Type	Required	Description
name	string	yes	Agent display name
slug	string	yes	URL-safe slug (e.g. lead-qualifier)
description	string	yes	What this agent does
system_prompt	string	yes	Agent system prompt — defines personality, expertise, and behavior
avatar_emoji	string	no
model	string	no
temperature	number	no
max_tokens	number	no
allowed_tools	string[]	no
denied_tools	string[]	no
custom_tool_ids	string[]	no
max_tool_iterations	number	no
context_instructions	string	no
triggers	object[]	no

update_agent

destructive

Update a subagent configuration. Can change prompt, tools, triggers, or model.

Parameter	Type	Required	Description
slug	string	yes	Agent slug to update
name	string	no
description	string	no
system_prompt	string	no
model	string	no
temperature	number	no
allowed_tools	string[]	no
denied_tools	string[]	no
custom_tool_ids	string[]	no
triggers	object[]	no
context_instructions	string	no
enabled	boolean	no

delete_agent

destructive

Delete a subagent from this workspace.

Parameter	Type	Required	Description
slug	string	yes	Agent slug to delete

invoke_agent

action

Invoke a subagent with a message. The agent runs with its own system prompt and tool access, then returns the result.

Parameter	Type	Required	Description
slug	string	yes	Agent slug to invoke
message	string	yes	Message to send to the agent
context	object	no

get_agent_memories

read-only

Get agent learnings and memories. These are data-driven insights the system has learned from campaign outcomes, reply patterns, and user feedback. Use this to understand what works for this workspace.

Parameter	Type	Required	Description
category	enum(campaign_learning \	icp_insight \	copy_pattern \
agent_id	string	no
min_confidence	number	no
limit	number	no

store_agent_memory

action

Store a new learning/insight for the agent. Use this when you discover something about the workspace, their ICP, messaging patterns, objections, or preferences.

Parameter	Type	Required	Description
category	enum(campaign_learning \	icp_insight \	copy_pattern \
content	string	yes	The insight to remember
confidence	number	no
agent_id	string	no
source	string	no

reinforce_memory

action

Reinforce an existing memory with new evidence. Increases confidence score.

Parameter	Type	Required	Description
memory_id	string	yes	Memory ID to reinforce
evidence	object	no

weaken_memory

action

Weaken a memory when contradicting evidence is found. Decreases confidence.

Parameter	Type	Required	Description
memory_id	string	yes	Memory ID to weaken

create_agent_chain

action

Create a chain of agents that run in sequence, passing outputs between them. Each agent can use different MCP tools (Trigify, CRM, Apollo, Email Outreach, etc.). Chains are saved as triggers and can run on schedule or on events. Beta feature — use for custom workflows that don’t fit the standard pipeline. IMPORTANT: Always call list_agents first to get real agent IDs. If agents don’t exist, create them with create_agent before creating the chain.

Parameter	Type	Required	Description
name	string	yes	Chain name (e.g., “Trigify → Enrich → Email Outreach”)
steps	object[]	yes	Ordered agent steps — minimum 2 agents
trigger_type	enum(schedule \	event \	manual)
schedule	string	no
event_name	string	no

run_agent_chain

action

Manually run an agent chain. Pass optional context data that the first agent receives. Set dry_run=true for a safe test run — opted-in side-effect tools (send_linkedin_message, add_leads_to_heyreach_campaign, bulk_enroll_whatsapp) will no-op and return their would-be payload.

Parameter	Type	Required	Description
chain_id	string	yes	Chain trigger ID (from create_agent_chain)
context	object	no
dry_run	boolean	no

list_agent_chains

read-only

List all agent chains in this workspace.

No parameters.

get_chain_history

read-only

Get execution history for agent chains — shows each run with steps, quality scores, tool calls, and any issues. Use this to debug and optimize chains. Reads the new agent_chain_runs tables first; falls back to audit_log for runs predating the migration.

Parameter	Type	Required	Description
chain_name	string	no
hours	number	no
limit	number	no

submit_chain_feedback

action

Submit human feedback on a chain step — used to improve agent performance. The feedback is stored in agent memory and influences future runs.

Parameter	Type	Required	Description
chain_id	string	yes	Chain execution ID (from get_chain_history)
step_index	number	yes	Which step to give feedback on (1-based)
rating	enum(good \	bad \	needs_improvement)
feedback	string	yes	What was wrong or right, what should change
agent_id	string	no

add_chain_step

action

Add a step to an existing chain. By default appends to the end; set insert_after to insert at a specific position (1-based).

Parameter	Type	Required	Description
chain_id	string	yes	Chain trigger ID
agent_id	string	yes	Workspace agent that will run this step
message_template	string	no
output_key	string	no
insert_after	number	no

delete_chain_step

destructive

Remove a step from a chain by 1-based position. Remaining steps shift down.

Parameter	Type	Required	Description
chain_id	string	yes
position	number	yes	1-based step position to remove

set_chain_step_condition

destructive

Attach (or clear) a condition on a chain step. The step only runs when the condition evaluates true against the accumulated chain context. Optionally jump to else_jump_to (1-based) when false. Pass condition: null to clear an existing condition.

Parameter	Type	Required	Description
chain_id	string	yes
position	number	yes	1-based step position
condition	object	no
else_jump_to	number	no

list_chain_templates

read-only

List battle-tested chain templates available for one-click install. Each template has a stable slug; pass it to install_chain_template to deploy.

No parameters.

install_chain_template

action

Install a battle-tested chain template into this workspace. Missing agents that the template references are auto-created from the agent template registry. Idempotent: re-installing the same template returns the existing chain id unless force=true.

Parameter	Type	Required	Description
template_slug	string	yes	Stable slug from list_chain_templates (e.g. “post-engagers-to-leads-view”)
name_override	string	no
force	boolean	no

set_chain_step_parallel_group

destructive

Tag a step with a parallel_group. Adjacent steps sharing the same group run concurrently via Promise.all. Pass parallel_group: null to make the step sequential again.

Parameter	Type	Required	Description
chain_id	string	yes
position	number	yes
parallel_group	string	no

update_chain_step

destructive

Edit message_template and/or output_key on an existing chain step. To change the agent on a step, delete it and add a new one (this preserves the audit trail).

Parameter	Type	Required	Description
chain_id	string	yes
position	number	yes
message_template	string	no
output_key	string	no

set_chain_enabled

destructive

Activate or deactivate a chain. When disabled, scheduled fires are skipped, event triggers do not match, and manual runs are still allowed (so users can test before re-enabling). Equivalent to the on/off switch on the chain detail page.

Parameter	Type	Required	Description
chain_id	string	yes
enabled	boolean	yes

set_chain_step_retry

destructive

Configure retry policy on a chain step. Transient failures (rate limits, 5xx, agent timeouts) are retried with exponential backoff before the step is recorded as failed. Set max_attempts: 1 (or pass null) to disable retries. Capped at 5 attempts and 30s backoff per attempt.

Parameter	Type	Required	Description
chain_id	string	yes
position	number	yes	1-based step position
retry	object	no
continue_on_failure	boolean	no

set_chain_max_concurrency

destructive

Set how many runs of this chain can be in flight at once. Default 1 — schedule fires that arrive while a previous run is still going are silently dropped. Set higher for chains that are genuinely safe to run in parallel (e.g. per-lead reply triage). Set 0 to disable the lock entirely (advanced).

Parameter	Type	Required	Description
chain_id	string	yes
max_concurrent_runs	number	yes

diagnose_chain_run

action

Self-debug a failed or low-quality chain run. Returns a structured diagnosis: which step failed, the error, the resolved prompt + actual output, and a list of suggested fixes (add retry, edit prompt, gate with a condition, swap agent). Use this when a user asks “why did my chain fail?” — the response gives Command Center the context to either explain the failure or apply a fix via set_chain_step_retry / update_chain_step / set_chain_step_condition.

Parameter	Type	Required	Description
run_id	string	yes	agent_chain_runs.id

Triggers

Event-driven automation triggers (events, schedules, keywords, thresholds)

list_triggers

read-only

List all automation triggers for this workspace. Triggers fire agents automatically based on events, schedules, keywords, or thresholds.

Parameter	Type	Required	Description
agent_id	string	no

create_trigger

action

Create a new automation trigger. Types:

• event: fires on system events (lead_replied, meeting_booked, etc.)
• schedule: fires on cron schedule
• keyword: fires when keyword detected in message
• threshold: fires when metric crosses threshold
• pipeline: fires at pipeline step completion

Parameter	Type	Required	Description
agent_id	string	yes	Agent to invoke when trigger fires
name	string	yes	Trigger name
trigger_type	enum(event \	schedule \	webhook \
config	object	yes	Trigger configuration (varies by type)
action_type	enum(invoke_agent \	send_notification \	update_record \
action_config	object	no
cooldown_minutes	number	no
max_fires_per_day	number	no

update_trigger

destructive

Update an existing trigger configuration.

Parameter	Type	Required	Description
trigger_id	string	yes	Trigger ID
name	string	no
config	object	no
action_config	object	no
enabled	boolean	no
cooldown_minutes	number	no
max_fires_per_day	number	no

delete_trigger

destructive

Delete an automation trigger.

Parameter	Type	Required	Description
trigger_id	string	yes	Trigger ID to delete

get_trigger_log

read-only

View recent trigger execution history. Shows when triggers fired, results, and errors.

Parameter	Type	Required	Description
trigger_id	string	no
limit	number	no

Calendar

Calendar integrations (Calendar, Calendar), booking events, meeting outcomes

list_calendar_integrations

read-only

List connected calendar integrations with sync status, last sync time, and error state. Use setup_calendar to connect a new calendar.

No parameters.

setup_calendar

action

Configure a calendar integration (Calendar or Calendar). Provide API key and booking URL to enable automatic meeting tracking.

Parameter	Type	Required	Description
provider	enum(cal_com \	calendly \	google_calendar \
api_key	string	no
booking_url	string	no
event_type_id	string	no
duration_minutes	number	no
sync_direction	enum(inbound \	outbound \	bidirectional)

sync_calendar

action

Sync calendar events from the configured provider. Pulls new bookings and updates lead statuses.

Parameter	Type	Required	Description
provider	enum(cal_com \	calendly \	google_calendar \

list_calendar_events

read-only

List calendar events (meetings) with their linked leads.

Parameter	Type	Required	Description
status	enum(scheduled \	confirmed \	cancelled \
upcoming_only	boolean	no
lead_id	string	no
limit	number	no

update_meeting_outcome

destructive

Record meeting outcome after it has occurred.

Parameter	Type	Required	Description
event_id	string	yes	Calendar event ID
status	enum(completed \	no_show \	cancelled)
notes	string	no
rating	number	no

Calendar Booking

Provider-agnostic check_calendar_availability + book_meeting that any agent can call. Dispatches to Calendar / Calendar via the adapter SDK.

check_calendar_availability

read-only

Returns ISO 8601 start times of bookable slots on a sender’s connected calendar (Calendar or Calendar) for the next N days. Use this BEFORE proposing or booking any specific time.

Parameter	Type	Required	Description
sender_id	string	yes	UUID of the sender whose calendar to check.
days_ahead	number	yes	How many days ahead to scan. Recommend 7. Max 14.
time_zone	string	no

book_meeting

action

Books a meeting on a sender’s connected calendar. Calendar confirms directly. Calendar returns either a confirmed booking (paid Scheduling API plan) or a one-time scheduling link (free plan) — caller should embed the link in the next outbound reply when kind=‘link_only’. Use ONLY a slot returned by check_calendar_availability.

Parameter	Type	Required	Description
sender_id	string	yes	UUID of the sender whose calendar to book.
start_iso	string	yes	ISO 8601 start time of the meeting.
attendee_name	string	yes	Lead’s full name.
attendee_email	string	yes	Lead’s email — required for the calendar invite.
attendee_phone	string	no
attendee_time_zone	string	no
attendee_language	enum(de \	en)	no
notes	string	no

get_booking_link

read-only

Returns the public scheduling link to send to a lead so they can self-book. Use this when the lead prefers picking their own time over confirming a specific slot. Prefers the per-sender link (senders.metadata.calendar_link); falls back to the workspace’s connected Calendar / Calendar booking URL — preferring whichever provider the sender is configured for.

Parameter	Type	Required	Description
sender_id	string	yes	UUID of the sender to look up.

Skills

Reusable agent capabilities (objection handler, meeting scheduler, etc.)

list_skills

read-only

List available skills for agents. Skills are reusable prompt+tool bundles that give agents specific capabilities (e.g., meeting-scheduler, objection-handler, campaign-optimizer).

Parameter	Type	Required	Description
scope	enum(org \	public \	builtin)
tags	string[]	no
search	string	no

get_agent_skills

read-only

Get skills assigned to a specific agent.

Parameter	Type	Required	Description
agent_id	string	yes	Agent ID

assign_skill

action

Assign a skill to an agent, giving it that capability.

Parameter	Type	Required	Description
agent_id	string	yes	Agent ID
skill_id	string	yes	Skill ID
config	object	no

remove_skill

destructive

Remove a skill from an agent.

Parameter	Type	Required	Description
agent_id	string	yes	Agent ID
skill_id	string	yes	Skill ID

add_skill_example

action

Add a training example for a skill. The agent will learn from these examples.

Parameter	Type	Required	Description
skill_slug	string	yes	Which skill to train (e.g. “objection-handler”)
example	string	yes	The training example (e.g. “When they say price is too high, respond with ROI calculation”)

create_skill

action

Create a new custom skill for this workspace.

Parameter	Type	Required	Description
name	string	yes	Skill name
slug	string	yes	URL-safe identifier
description	string	yes	What this skill does
system_prompt_extension	string	yes	Instructions appended to agent prompt when skill is active
required_tools	string[]	no
trigger_patterns	string[]	no
trigger_events	string[]	no
tags	string[]	no
is_public	boolean	no

Background Jobs

Queue and track long-running jobs in the background worker

enqueue_job

action

Queue a long-running job to the background worker. Use for tasks that take >5 minutes: pipeline runs with many companies, deep research on 50+ leads, long agent chains, bulk operations. Returns a job_id for tracking.

Parameter	Type	Required	Description
type	enum(pipeline_run \	deep_research \	agent_chain \
payload	object	yes	Job-specific parameters. For pipeline_run: { playbook_id, limit }. For deep_research: { company_ids } or { company_names }. For agent_chain: { chain_id, input }. For playbook_optimize: { playbook_id, auto_apply }. For bulk_outreach: { lead_ids, sequence_id }.
priority	number	no

get_job_status

read-only

Check the status of a background job by ID.

Parameter	Type	Required	Description
job_id	string	yes

list_jobs

read-only

List recent background jobs (imports, bulk actions, pipeline runs) with their status and progress.

Parameter	Type	Required	Description
status	enum(pending \	processing \	done \
limit	number	no

cancel_job

action

Cancel a pending background job. Cannot cancel running jobs.

Parameter	Type	Required	Description
job_id	string	yes

bulk_agent_run

action

Run a workspace agent on multiple companies in the background. The prompt template supports placeholders: {{company_name}}, {{domain}}, {{industry}}, {{company_id}}, {{location}}, {{employee_count}}. Returns a batch_id for tracking progress. Use get_batch_status to monitor.

Parameter	Type	Required	Description
agent_id	string	yes	Workspace agent ID (get from list_agents)
prompt_template	string	yes	Message template for the agent. Use {{company_name}}, {{domain}}, etc.
company_ids	string[]	yes	Company IDs to process
playbook_id	string	no

get_batch_status

read-only

Check the status and progress of a pipeline batch (enrichment, agent run, etc.).

Parameter	Type	Required	Description
batch_id	string	yes

list_active_batches

read-only

List active pipeline batches (running enrichments, agent runs, etc.) for a playbook.

Parameter	Type	Required	Description
playbook_id	string	yes

Autonomous Missions

Launch, monitor, and approve long-running autonomous missions

create_mission

action

Launch an autonomous mission: a long-running agent that plans and works toward a goal on the background worker, using read-only workspace tools (research, lead/company/pipeline analytics). It CANNOT send, enrol, or book — it pauses and asks for approval, and the platform executes any approved action. Use for open-ended work like “analyse why playbook X underperforms and draft a fix” or “research 20 lookalikes of our converts and propose a segment”. Returns a mission_id; poll with get_mission.

Parameter	Type	Required	Description
goal	string	yes	What the mission should accomplish. Be specific about the deliverable.

list_missions

read-only

List this workspace’s autonomous missions (newest first) with their status. Filter by status to find ones awaiting approval.

Parameter	Type	Required	Description
status	enum(running \	waiting_approval \	done \
limit	number	no

get_mission

read-only

Get a mission’s full state: goal, status, living plan, recent event trace, registered deliverables (artifacts), and — if it is waiting_approval — exactly what it is asking to do so you can decide_mission.

Parameter	Type	Required	Description
mission_id	string	yes	Mission ID from create_mission or list_missions.

decide_mission

action

Resolve a mission that is waiting_approval: approve, reject, or request edits to what it proposed. Approving an action lets the PLATFORM execute it (e.g. the proposed outbound step) and resumes the mission; reject/edit resumes it with your guidance. Only valid when the mission status is waiting_approval.

Parameter	Type	Required	Description
mission_id	string	yes
decision	enum(approve \	reject \	edit)
note	string	no