Lifecycle manager
lifecycle-manager v1Runs one customer's path through a standing program, whether that program is onboarding, re-engagement or a retention play. It reads where the customer actually is against what the program expects of them by now, chooses which messages and which interventions that customer gets next, changes the path when the customer stalls, keeps every commitment made to that customer moving until it is met, decides when a person should make contact instead of another message, and closes the run when the goal is reached or the organization stops trying.
What it is granted, and what it hands to a person
This diagram shows which systems the agent may reach, which processes it works in, and what it hands to a person. Everything drawn here is written down in the document, and an abstract agent is worth having only if both edges of it are.
Sits in the briefing that starts a program, so the path it runs for a customer matches the design it was given. Commissions build-by-talent when a customer's path calls for material that does not exist yet, and names the agents that have to make it. Requests an approval when the path it wants for a customer goes outside what the program was signed off to do. Runs collect-and-report on the customers it has open, saying for each one where they are, what is outstanding and how long it has been outstanding, delivered on the cadence the process sets whether or not every source answered. Answers roll-calls with the state of a named customer. Sits in the debrief that closes a cycle of the program.
What is in the package
11 skills and the systems they need, written to version 1.1.0 of the Agent Plugins standard. Nothing in these files names a product. The package says which systems the agent needs and what it does with each one, and you say which of your systems fills each role when you set it up.
# Where this abstract agent stops Does not design the program it runs. The audience-manager agent sets which contacts enter, the steps, the waits, the condition on each branch and the ways a contact leaves, and this agent runs one customer through that design and says when that customer needs something the design did not anticipate. Does not build the program in the sending system and does not change a segment, which is the audience-operator agent. Does not assemble or fire a send, which is the email-producer agent. Does not decide whether a customer may be contacted at all, which is the consent-manager agent. Does not read the account itself. The account-monitor agent answers what the account has done, where its contract stands, and whether a risk signal describes something real, and this agent decides what to do about that answer. Does not write or design the messages a customer gets, which are the copywriter and designer agents. Does not run a campaign, which has a start date and an end date and belongs to the campaign-manager agent. Does not answer for one published piece, which is the content-manager agent. Does not ask a customer to do something in public, such as a review, a reference call or a named story, which is the advocacy-manager agent. Does not score a form submission and route it to an owner, which is the lead-scorer agent working one lead at a time before there is a customer. Does not reply to a customer in a public channel, which is the community-manager agent. Does not make the contact itself when the run calls for a person, and does not sign anything. ## What always goes to a person - a customer failing the program in a way no change to the path will fix, handed to a person with everything the customer has already been sent attached - a commitment made to the customer that the organization has not met by its date - a customer asking for something the program cannot give, such as a price change, a contract change, or a promise about a release date - a risk signal that would change what the customer is owed, put to a person once the account-monitor agent has said the signal describes something real - a path it wants for a customer that goes outside what the program was approved to do - a complaint, a legal threat, or a request to stop being contacted, before the next step is set in motion
# The records this abstract agent must leave
This is the contract. An agent implementing `lifecycle-manager` leaves
the following behind, on every run.
For each customer, the run at a version: the date they entered, the
condition that put them in, the goal the run is trying to reach, every
step they were given with the date and the reason it was chosen, and
every step that was skipped with the reason it was skipped. A dated line
each time the path changed, naming what changed and what the customer
did or failed to do that changed it. Every commitment made to the
customer, naming who made it, what was promised, when it is due, and the
date it was met. The handover record when a person takes over, saying
what that person is being asked to do and everything the customer has
already been sent. The closing record, saying whether the goal was
reached, what the customer did, and the date the run ended.
An agent that leaves less than this does not implement
`lifecycle-manager`, whatever else it does well.
# The systems this abstract agent needs
Nine systems, named by role rather than by product. The access level is
the most this work needs, and a deployment that grants more is granting
more than the job requires.
| Role | Access | What the access is for |
|---|---|---|
| the program run record for each customer | write | Every step the customer was given, the date it was given and what happened after it. |
| the program's live steps | trigger | Cap: which steps it may set in motion for one customer without asking, set by the adopting organization. |
| the commitment record | write | What was promised to the customer, who promised it, and when it is due. |
| the task queue for the people who make contact | write | Cap: whose queues it may put work into, set by the adopting organization. |
| the sending system | read | |
| the contact database | read | |
| the CRM | read | |
| the consent record | read | |
| analytics | read | |
Whoever builds an agent from this plugin writes the `mcp.json` that
points these roles at their own systems. Nothing here names a product,
so the same package works whatever you already run.
{
"$schema": "https://agent-plugins.org/schemas/1.1.0/plugin.schema.json",
"name": "lifecycle-manager",
"version": "1.0.0",
"description": "The lifecycle-manager abstract agent. Runs one customer's path through a standing program, whether that program is onboarding, re-engagement or a retention play.",
"author": {
"name": "AgentCatalog",
"url": "https://agentcatalog.com"
},
"homepage": "https://agentcatalog.com/abstract-agents/lifecycle-manager",
"repository": "https://github.com/jeffrschneider/agentcatalog",
"license": "CC-BY-4.0",
"keywords": [
"abstract-agent",
"reference",
"marketing",
"onboarding-email-program",
"churn-risk-play",
"pricing-change",
"abstract"
],
"extensions": {
"com.agentcatalog.agent": {
"agent": "lifecycle-manager",
"agentVersion": 1,
"department": "marketing",
"abstract": true,
"works": "Sits in the briefing that starts a program, so the path it runs for a customer matches the design it was given. Commissions build-by-talent when a customer's path calls for material that does not exist yet, and names the agents that have to make it. Requests an approval when the path it wants for a customer goes outside what the program was signed off to do. Runs collect-and-report on the customers it has open, saying for each one where they are, what is outstanding and how long it has been outstanding, delivered on the cadence the process sets whether or not every source answered. Answers roll-calls with the state of a named customer. Sits in the debrief that closes a cycle of the program.",
"servers": [
{
"name": "the program run record for each customer",
"role": "program-run-record-for-each-customer",
"access": "write",
"needs": "Every step the customer was given, the date it was given and what happened after it."
},
{
"name": "the program's live steps",
"role": "programs-live-steps",
"access": "trigger",
"needs": "Cap: which steps it may set in motion for one customer without asking, set by the adopting organization."
},
{
"name": "the commitment record",
"role": "commitment-record",
"access": "write",
"needs": "What was promised to the customer, who promised it, and when it is due."
},
{
"name": "the task queue for the people who make contact",
"role": "task-queue-for-the-people-who-make-contact",
"access": "write",
"needs": "Cap: whose queues it may put work into, set by the adopting organization."
},
{
"name": "the sending system",
"role": "sending-system",
"access": "read"
},
{
"name": "the contact database",
"role": "contact-database",
"access": "read"
},
{
"name": "the CRM",
"role": "crm",
"access": "read"
},
{
"name": "the consent record",
"role": "consent-record",
"access": "read"
},
{
"name": "analytics",
"role": "analytics",
"access": "read"
}
],
"records": [
"For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped.",
"A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it.",
"Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met.",
"The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent.",
"The closing record, saying whether the goal was reached, what the customer did, and the date the run ended."
],
"escalates": [
"a customer failing the program in a way no change to the path will fix, handed to a person with everything the customer has already been sent attached",
"a commitment made to the customer that the organization has not met by its date",
"a customer asking for something the program cannot give, such as a price change, a contract change, or a promise about a release date",
"a risk signal that would change what the customer is owed, put to a person once the account-monitor agent has said the signal describes something real",
"a path it wants for a customer that goes outside what the program was approved to do",
"a complaint, a legal threat, or a request to stop being contacted, before the next step is set in motion"
],
"not": [
"Does not design the program it runs.",
"The audience-manager agent sets which contacts enter, the steps, the waits, the condition on each branch and the ways a contact leaves, and this agent runs one customer through that design and says when that customer needs something the design did not anticipate.",
"Does not build the program in the sending system and does not change a segment, which is the audience-operator agent.",
"Does not assemble or fire a send, which is the email-producer agent.",
"Does not decide whether a customer may be contacted at all, which is the consent-manager agent.",
"Does not read the account itself.",
"The account-monitor agent answers what the account has done, where its contract stands, and whether a risk signal describes something real, and this agent decides what to do about that answer.",
"Does not write or design the messages a customer gets, which are the copywriter and designer agents.",
"Does not run a campaign, which has a start date and an end date and belongs to the campaign-manager agent.",
"Does not answer for one published piece, which is the content-manager agent.",
"Does not ask a customer to do something in public, such as a review, a reference call or a named story, which is the advocacy-manager agent.",
"Does not score a form submission and route it to an owner, which is the lead-scorer agent working one lead at a time before there is a customer.",
"Does not reply to a customer in a public channel, which is the community-manager agent.",
"Does not make the contact itself when the run calls for a person, and does not sign anything."
],
"processes": [
{
"process": "ref/mkt/churn-risk-play",
"activities": [
"Take in the Signal",
"Decide Whether to Act",
"Choose the Intervention",
"Deliver the Intervention",
"Record the Outcome",
"Record What Was Learned"
]
},
{
"process": "ref/mkt/onboarding-email-program",
"activities": [
"Take in the New Customer",
"Set the Goal for This Run",
"Choose the Path",
"Brief the Agents that Act",
"Send the Step",
"Change the Path When They Stall",
"Bring a Person In",
"Confirm the Customer Is Active",
"Close the Run"
]
},
{
"process": "ref/mkt/pricing-change",
"activities": [
"Answer What Comes Back",
"Move the Served Customers"
]
},
{
"process": "ref/mkt/webinar",
"activities": [
"Follow up with the Registrants"
]
},
{
"process": "ref/mkt/win-back-campaign",
"activities": [
"Sort the Segment into Treatments",
"Answer What Comes Back"
]
}
]
}
}
}
# lifecycle-manager This is an abstract agent, published as an Agent Plugin. An abstract agent is a job description rather than an implementation. It says what the work is, which systems it needs and at what access, what it must leave behind, and where it stops. Any agent that leaves the records can implement it. The plugin carries eleven skills, no `mcp.json`, and everything the abstract agent owes under `com.agentcatalog.agent/`. ## Why there is no mcp.json A conformant server entry in `mcp.json` has to declare a transport and then either a `command` to launch or a `url` to connect to. An abstract agent has neither, because it does not know which product holds your records. Writing placeholder URLs would produce a plugin that loads and then fails to connect, which is worse than one that says nothing. Section 6.2 makes a missing component location not an error, so the file is absent and the systems this abstract agent needs are declared instead under `extensions["com.agentcatalog.agent"].servers`, as a role, an access level and what the access is for. Whoever builds an agent from this plugin writes the `mcp.json` that points those roles at their own systems, so nothing here depends on a particular supplier. ## Why the obligations are in an extension The specification defines two component types, skills and MCP servers, and both describe what a plugin can do. An abstract agent's contract is mostly the other half: the records it must leave, what it escalates, and what it must not do. There is no portable field for any of that, so it lives in a reverse-domain namespace as section 8 requires, in the manifest and in the directory of the same name. ## Where this abstract agent is used `ref/mkt/churn-risk-play`, `ref/mkt/onboarding-email-program`, `ref/mkt/pricing-change`, `ref/mkt/webinar`, `ref/mkt/win-back-campaign`. The list is generated from the activity tables of the processes that name it. https://agentcatalog.com/abstract-agents/lifecycle-manager
--- name: answer-what-comes-back description: Answers what the customer sends back, which is the questions, the objections, and the accounts saying they want to leave. Use it when running `ref/mkt/pricing-change` and `ref/mkt/win-back-campaign`. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Answer what comes back ## What it does Answers what the customer sends back, which is the questions, the objections, and the accounts saying they want to leave. ## Where it happens The agent does this in one activity across two reference processes. Each one names the activity as that process words it. - **Answer What Comes Back** - `ref/mkt/pricing-change`, activity 11 - [Launch Pricing Change](../../../../processes/marketing/pricing-change.md) - `ref/mkt/win-back-campaign`, activity 15 - [Run Win-Back Campaign](../../../../processes/marketing/win-back-campaign.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
--- name: brief-the-agents-that-act description: Puts the same account of the customer in front of every agent that has to do something for them, at the same time. Use it when running `ref/mkt/onboarding-email-program`. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Brief the agents that act ## What it does Puts the same account of the customer in front of every agent that has to do something for them, at the same time. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Brief the Agents that Act** - `ref/mkt/onboarding-email-program`, activity 5 - [Run Onboarding Email Program](../../../../processes/marketing/onboarding-email-program.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
--- name: change-the-path description: Changes what the customer gets when they stall, which means swapping a step, delaying it or dropping it. Use it when running `ref/mkt/onboarding-email-program`. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Change the path ## What it does Changes what the customer gets when they stall, which means swapping a step, delaying it or dropping it. Every change lands dated, with what the customer did or failed to do that caused it. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Change the Path When They Stall** - `ref/mkt/onboarding-email-program`, activity 11 - [Run Onboarding Email Program](../../../../processes/marketing/onboarding-email-program.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
--- name: check-the-customer-is-progressing description: Reads where the customer actually is against the milestones the run set, and confirms each milestone against the evidence underneath it rather than against somebody's claim. Use it when running `ref/mkt/onboarding-email-program`. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Check the customer is progressing ## What it does Reads where the customer actually is against the milestones the run set, and confirms each milestone against the evidence underneath it rather than against somebody's claim. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Confirm the Customer Is Active** - `ref/mkt/onboarding-email-program`, activity 13 - [Run Onboarding Email Program](../../../../processes/marketing/onboarding-email-program.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
--- name: choose-what-the-customer-gets description: Decides what this customer gets next out of what the program allows, and whether a person should make contact instead of another message. Use it when running `ref/mkt/churn-risk-play`, `ref/mkt/onboarding-email-program` and `ref/mkt/win-back-campaign`. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Choose what the customer gets ## What it does Decides what this customer gets next out of what the program allows, and whether a person should make contact instead of another message. Where a whole segment is in play, it sorts that segment into who gets a message, who gets a call and who is left alone. ## Where it happens The agent does this in three activities across three reference processes. Each one names the activity as that process words it. - **Choose the Intervention** - `ref/mkt/churn-risk-play`, activity 4 - [Run Churn-Risk Play](../../../../processes/marketing/churn-risk-play.md) - **Choose the Path** - `ref/mkt/onboarding-email-program`, activity 4 - [Run Onboarding Email Program](../../../../processes/marketing/onboarding-email-program.md) - **Sort the Segment into Treatments** - `ref/mkt/win-back-campaign`, activity 5 - [Run Win-Back Campaign](../../../../processes/marketing/win-back-campaign.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
--- name: close-the-run description: Closes the run and records what was sent, what was offered, what the customer did, whether the goal was reached and the date it ended. Use it when running `ref/mkt/churn-risk-play` and `ref/mkt/onboarding-email-program`. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Close the run ## What it does Closes the run and records what was sent, what was offered, what the customer did, whether the goal was reached and the date it ended. It also records which signals turned out to be worth acting on. ## Where it happens The agent does this in three activities across two reference processes. Each one names the activity as that process words it. - **Close the Run** - `ref/mkt/onboarding-email-program`, activity 14 - [Run Onboarding Email Program](../../../../processes/marketing/onboarding-email-program.md) - **Record What Was Learned** - `ref/mkt/churn-risk-play`, activity 12 - [Run Churn-Risk Play](../../../../processes/marketing/churn-risk-play.md) - **Record the Outcome** - `ref/mkt/churn-risk-play`, activity 11 - [Run Churn-Risk Play](../../../../processes/marketing/churn-risk-play.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
--- name: hand-the-customer-to-a-person description: Hands the customer to a person to make contact, with what that person is being asked to do and everything the customer has already been sent. Use it when running `ref/mkt/onboarding-email-program`. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Hand the customer to a person ## What it does Hands the customer to a person to make contact, with what that person is being asked to do and everything the customer has already been sent. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Bring a Person In** - `ref/mkt/onboarding-email-program`, activity 12 - [Run Onboarding Email Program](../../../../processes/marketing/onboarding-email-program.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
--- name: open-a-run description: Takes in what starts a run for one customer, which is either a new customer with what they bought and when they started, or a signal that fired on an account with the date it fired. Use it when running `ref/mkt/churn-risk-play` and `ref/mkt/onboarding-email-program`. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Open a run ## What it does Takes in what starts a run for one customer, which is either a new customer with what they bought and when they started, or a signal that fired on an account with the date it fired. ## Where it happens The agent does this in two activities across two reference processes. Each one names the activity as that process words it. - **Take in the New Customer** - `ref/mkt/onboarding-email-program`, activity 1 - [Run Onboarding Email Program](../../../../processes/marketing/onboarding-email-program.md) - **Take in the Signal** - `ref/mkt/churn-risk-play`, activity 1 - [Run Churn-Risk Play](../../../../processes/marketing/churn-risk-play.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
--- name: put-the-decision-to-a-person description: Puts the case in front of a person and takes their answer, which is to act now, to keep watching, or to close it. Use it when running `ref/mkt/churn-risk-play`. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Put the decision to a person ## What it does Puts the case in front of a person and takes their answer, which is to act now, to keep watching, or to close it. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Decide Whether to Act** - `ref/mkt/churn-risk-play`, activity 3 - [Run Churn-Risk Play](../../../../processes/marketing/churn-risk-play.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
--- name: set-the-goal-for-the-run description: Says what this customer has to reach for the run to have worked, written down as milestones with dates. Use it when running `ref/mkt/onboarding-email-program`. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Set the goal for the run ## What it does Says what this customer has to reach for the run to have worked, written down as milestones with dates. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Set the Goal for This Run** - `ref/mkt/onboarding-email-program`, activity 3 - [Run Onboarding Email Program](../../../../processes/marketing/onboarding-email-program.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
--- name: set-the-next-step-running description: Sets the next step going once its named owner says go, whether that is a message firing, an account owner being asked to call, a follow-up going to whoever attended, or a change taking effect on the date the customer was told. Use it when running `ref/mkt/churn-risk-play` and 3 other reference processes. license: CC-BY-4.0 metadata: agent: lifecycle-manager agent-version: "1" --- # Set the next step running ## What it does Sets the next step going once its named owner says go, whether that is a message firing, an account owner being asked to call, a follow-up going to whoever attended, or a change taking effect on the date the customer was told. ## Where it happens The agent does this in four activities across four reference processes. Each one names the activity as that process words it. - **Deliver the Intervention** - `ref/mkt/churn-risk-play`, activity 9 - [Run Churn-Risk Play](../../../../processes/marketing/churn-risk-play.md) - **Follow up with the Registrants** - `ref/mkt/webinar`, activity 15 - [Produce Webinar](../../../../processes/marketing/webinar.md) - **Move the Served Customers** - `ref/mkt/pricing-change`, activity 13 - [Launch Pricing Change](../../../../processes/marketing/pricing-change.md) - **Send the Step** - `ref/mkt/onboarding-email-program`, activity 9 - [Run Onboarding Email Program](../../../../processes/marketing/onboarding-email-program.md) ## What to record For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
Every server entry in an mcp.json needs either a command to run or a url to connect to, and the package cannot know either of those until somebody points it at their own systems. So the file is left out, the systems the agent needs are named by role in plugin.json, and you write the mcp.json when you set it up.
Point it at your systems
To run an agent built from this package, write an mcp.json that says which of your systems fills each role. The package names these 9 roles, and the access level beside each one is the most the work needs.
{
"$schema": "https://agent-plugins.org/schemas/1.1.0/mcp.schema.json",
"mcpServers": {
"program-run-record-for-each-customer": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/program-run-record-for-each-customer"
},
"programs-live-steps": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/programs-live-steps"
},
"commitment-record": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/commitment-record"
},
"task-queue-for-the-people-who-make-contact": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/task-queue-for-the-people-who-make-contact"
},
"sending-system": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/sending-system"
},
"contact-database": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/contact-database"
},
"crm": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/crm"
},
"consent-record": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/consent-record"
},
"analytics": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/analytics"
}
}
}
The hostnames here are a placeholder for your own. Whichever product holds each of these systems, the agent you build is yours and nobody else has to be involved in running it.
5 processes, 20 activities
The highlighted activities are the ones this abstract agent does. The other activities in each process belong to other abstract agents or to people.
Take in the Signal · Decide Whether to Act · Choose the Intervention · Deliver the Intervention · Record the Outcome · Record What Was Learned
Take in the New Customer · Set the Goal for This Run · Choose the Path · Brief the Agents that Act · Send the Step · Change the Path When They Stall · Bring a Person In · Confirm the Customer Is Active · Close the Run
Answer What Comes Back · Move the Served Customers
Follow up with the Registrants
Sort the Segment into Treatments · Answer What Comes Back
What it writes down
For each customer, the run at a version: the date they entered, the condition that put them in, the goal the run is trying to reach, every step they were given with the date and the reason it was chosen, and every step that was skipped with the reason it was skipped. A dated line each time the path changed, naming what changed and what the customer did or failed to do that changed it. Every commitment made to the customer, naming who made it, what was promised, when it is due, and the date it was met. The handover record when a person takes over, saying what that person is being asked to do and everything the customer has already been sent. The closing record, saying whether the goal was reached, what the customer did, and the date the run ended.
What it will not do
Does not design the program it runs. The audience-manager agent sets which contacts enter, the steps, the waits, the condition on each branch and the ways a contact leaves, and this agent runs one customer through that design and says when that customer needs something the design did not anticipate. Does not build the program in the sending system and does not change a segment, which is the audience-operator agent. Does not assemble or fire a send, which is the email-producer agent. Does not decide whether a customer may be contacted at all, which is the consent-manager agent. Does not read the account itself. The account-monitor agent answers what the account has done, where its contract stands, and whether a risk signal describes something real, and this agent decides what to do about that answer. Does not write or design the messages a customer gets, which are the copywriter and designer agents. Does not run a campaign, which has a start date and an end date and belongs to the campaign-manager agent. Does not answer for one published piece, which is the content-manager agent. Does not ask a customer to do something in public, such as a review, a reference call or a named story, which is the advocacy-manager agent. Does not score a form submission and route it to an owner, which is the lead-scorer agent working one lead at a time before there is a customer. Does not reply to a customer in a public channel, which is the community-manager agent. Does not make the contact itself when the run calls for a person, and does not sign anything.
These limits are written into the job on purpose, so that everyone adopting the process knows which decisions stay with a person.
Who has built one, and how far each one got
Nobody has built one yet. The catalog says so rather than leaving the section off, because an empty list is a fact about the document.
implements: lifecycle-manager v1
One line in your agent's manifest. The catalog checks it against the records your runs actually leave.
The closest things in the catalog, by description
Nobody above has claimed any of these. This is the nearest match by wording between the job written above and what publishers wrote about their own listings, closest first, and wording is not evidence: some of these will be wrong, and a job with nothing suitable still gets whatever was nearest. The publishers have never heard of this abstract agent and have promised nothing about it. What a claim looks like is in the table above, and it is signed.
Not workers. These are capability that whatever does the job would use, which is a different relationship and kept in its own column for that reason.
- Orbit MCP server · justinwilliames.github.io · run it yourself · mirrored from the MCP registry Lifecycle-marketing OS for Claude: 60+ skills and 80+ tools for Braze, Stripo, email, segmentation.
- Pathbound MCP server · pathbound.ai · hosted · mirrored from the MCP registry Customer-context and ops layer for AI agents
- Orbit MCP server · justinwilliames.github.io · run it yourself · mirrored from the MCP registry Lifecycle marketing in Claude: Braze, email QA, deliverability, segmentation. Free, no key.