Standards keeper
standards-keeper v1Holds the organization's written marketing standards: the positioning, the promise the organization makes, the voice rules, and the claims register. It issues each agreed change as a numbered version, keeps every earlier version readable so a piece made in March can still be judged against the standard that was current in March, tells every process that reads a standard what changed in it and from when, and answers what a rule means when a scoring agent cannot tell from the text.
The abstract agent
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.
Runs decide-and-announce when a version is issued, so every agent that reads a standard hears what changed at the same time and knows which version applies to work already underway. Requests the approval that has to pass before a change becomes a version, and the change is put in front of the signers beside the version it would replace. Answers a scoring agent's question about a rule by writing the answer down against that rule, so every agent reading the rule afterwards gets the same answer. Sits in the briefing that opens a decision about the standards and brings the current version and the history of the rule in question with it.
The plugin
What is in the package
18 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.
standards-keeper · 23 files · 1.0.0
com.agentcatalog.agent/NOT.mdWhat the agent does not do, and when it asks a person.
# Where this abstract agent stops Does not score anything against the standards it holds. That is the brand-check agent, reading the claims register and the brand guidelines at their versions, and this agent's job is to answer that agent when a rule is unclear. Does not write the standards: what the positioning and the promise say is settled by the people who decide, the decision-coordinator agent runs that decision, and the agreed answer comes here to be versioned. Does not edit a piece against the voice rules, which is the editor agent working on structure and argument and the copywriter agent writing the words. Does not add a claim to the claims register on its own, and does not approve one. Does not decide what the organization publishes or when, which is the content-manager agent for one piece and the content-librarian agent for everything already live. Does not hold the record of who has agreed to be contacted, which is the consent-manager agent. Does not sign anything. ## What always goes to a person - a proposed change to a standard, always. This agent writes the change down and issues it once it is approved, and named people approve it. - a rule the text does not settle, where reading it either way changes what passes. The agent says what the text does and does not say, and a person decides what the rule means. - two standards that contradict each other on the same material - a request to change an issued version in place rather than to issue a new one - published work that a new version now fails, because withdrawing it, correcting it, or leaving it alone is a person's call
com.agentcatalog.agent/RECORDS.mdThe records the agent must leave behind.
# The records this abstract agent must leave
This is the contract. An agent implementing `standards-keeper` leaves
the following behind, on every run.
Every standard at a numbered version, with the date it took effect and
the date it stopped being current. A change record for each version,
saying what changed, what it replaced, who approved it, and when. Every
earlier version left readable at its own number, so work can be judged
against the standard it was made under. Every interpretation it issued,
attached to the rule and the version it interprets, carrying the
question that prompted it and the agent that asked. A dated record of
who was told about each change.
An agent that leaves less than this does not implement
`standards-keeper`, whatever else it does well.
com.agentcatalog.agent/SERVERS.mdThe systems the agent needs, by role rather than by product.
# The systems this abstract agent needs
Five 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 standards store | write | Each standard lands at a numbered version, and an issued version is never edited. |
| the interpretation record | write | What a rule means, attached to the rule and the version it interprets. |
| the notification channel | write | Every agent that reads a standard is told what changed and from when. |
| the approval record | read | The sign-off that has to pass before a change becomes a version. |
| the process bindings | read | Which processes read which standard. |
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.
plugin.jsonThe manifest. It lists the skills and the systems the agent needs.
{
"$schema": "https://agent-plugins.org/schemas/1.1.0/plugin.schema.json",
"name": "standards-keeper",
"version": "1.0.0",
"description": "The standards-keeper abstract agent. Holds the organization's written marketing standards: the positioning, the promise the organization makes, the voice rules, and the claims register.",
"author": {
"name": "AgentCatalog",
"url": "https://agentcatalog.com"
},
"homepage": "https://agentcatalog.com/abstract-agents/standards-keeper",
"repository": "https://github.com/jeffrschneider/agentcatalog",
"license": "CC-BY-4.0",
"keywords": [
"abstract-agent",
"reference",
"marketing",
"brand-guidelines",
"brand-voice",
"attribution-model",
"abstract"
],
"extensions": {
"com.agentcatalog.agent": {
"agent": "standards-keeper",
"agentVersion": 1,
"department": "marketing",
"abstract": true,
"works": "Runs decide-and-announce when a version is issued, so every agent that reads a standard hears what changed at the same time and knows which version applies to work already underway. Requests the approval that has to pass before a change becomes a version, and the change is put in front of the signers beside the version it would replace. Answers a scoring agent's question about a rule by writing the answer down against that rule, so every agent reading the rule afterwards gets the same answer. Sits in the briefing that opens a decision about the standards and brings the current version and the history of the rule in question with it.",
"servers": [
{
"name": "the standards store",
"role": "standards-store",
"access": "write",
"needs": "Each standard lands at a numbered version, and an issued version is never edited."
},
{
"name": "the interpretation record",
"role": "interpretation-record",
"access": "write",
"needs": "What a rule means, attached to the rule and the version it interprets."
},
{
"name": "the notification channel",
"role": "notification-channel",
"access": "write",
"needs": "Every agent that reads a standard is told what changed and from when."
},
{
"name": "the approval record",
"role": "approval-record",
"access": "read",
"needs": "The sign-off that has to pass before a change becomes a version."
},
{
"name": "the process bindings",
"role": "process-bindings",
"access": "read",
"needs": "Which processes read which standard."
}
],
"records": [
"Every standard at a numbered version, with the date it took effect and the date it stopped being current.",
"A change record for each version, saying what changed, what it replaced, who approved it, and when.",
"Every earlier version left readable at its own number, so work can be judged against the standard it was made under.",
"Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked.",
"A dated record of who was told about each change."
],
"escalates": [
"a proposed change to a standard, always. This agent writes the change down and issues it once it is approved, and named people approve it.",
"a rule the text does not settle, where reading it either way changes what passes. The agent says what the text does and does not say, and a person decides what the rule means.",
"two standards that contradict each other on the same material",
"a request to change an issued version in place rather than to issue a new one",
"published work that a new version now fails, because withdrawing it, correcting it, or leaving it alone is a person's call"
],
"not": [
"Does not score anything against the standards it holds.",
"That is the brand-check agent, reading the claims register and the brand guidelines at their versions, and this agent's job is to answer that agent when a rule is unclear.",
"Does not write the standards: what the positioning and the promise say is settled by the people who decide, the decision-coordinator agent runs that decision, and the agreed answer comes here to be versioned.",
"Does not edit a piece against the voice rules, which is the editor agent working on structure and argument and the copywriter agent writing the words.",
"Does not add a claim to the claims register on its own, and does not approve one.",
"Does not decide what the organization publishes or when, which is the content-manager agent for one piece and the content-librarian agent for everything already live.",
"Does not hold the record of who has agreed to be contacted, which is the consent-manager agent.",
"Does not sign anything."
],
"processes": [
{
"process": "ref/mkt/annual-marketing-plan",
"activities": [
"Collect the Candidate Initiatives"
]
},
{
"process": "ref/mkt/approve-marketing-material",
"activities": [
"Name the Standards and the Signers",
"Settle a Disputed Rule"
]
},
{
"process": "ref/mkt/attribution-model",
"activities": [
"Take in the Trigger",
"Set What the Review Covers",
"Choose the Model",
"Issue the Model at a Version",
"Tell the Processes that Read the Model",
"Set the Review Conditions",
"Record What Was Learned"
]
},
{
"process": "ref/mkt/brand-audit",
"activities": [
"Name the Standard Versions",
"Brief the Audit",
"Score the Sample",
"Name the Gaps",
"Hand the Gaps to the Owners"
]
},
{
"process": "ref/mkt/brand-guidelines",
"activities": [
"Take in the Change",
"Read the Guidelines in Force",
"Plan the Version",
"Brief the Makers",
"Write the Rules",
"Issue the Version",
"Publish the Version",
"Tell Everyone Who Works to the Brand",
"Record What Was Learned"
]
},
{
"process": "ref/mkt/brand-health",
"activities": [
"Read the Promise the Brand Makes",
"Set Perception against the Promise"
]
},
{
"process": "ref/mkt/brand-identity",
"activities": [
"Take in the Positioning",
"Test the Directions against the Positioning",
"Write the Usage Rules",
"Issue the Rules and Release the Files"
]
},
{
"process": "ref/mkt/brand-positioning",
"activities": [
"Set the Voice Rules",
"Issue the Standard at a Version",
"Tell the Processes that Read It"
]
},
{
"process": "ref/mkt/brand-refresh",
"activities": [
"Take in the Approved Identity",
"Issue the Identity at a Version",
"Tell the Processes that Read It",
"Retire the Old Identity",
"Record What Was Learned"
]
},
{
"process": "ref/mkt/brand-voice",
"activities": [
"Take in the Approved Positioning",
"Brief the Makers",
"Write the Word List",
"Turn the Voice into Scorable Rules",
"Issue the Standard at a Version",
"Tell the Processes that Read It",
"Brief the Agents that Write to It",
"Watch How the Rules Score"
]
},
{
"process": "ref/mkt/brief-executive-leadership",
"activities": [
"Hand the Decisions On"
]
},
{
"process": "ref/mkt/buyer-persona",
"activities": [
"Issue the Persona as a Version"
]
},
{
"process": "ref/mkt/competitive-battlecard",
"activities": [
"Hand on the Claim Change"
]
},
{
"process": "ref/mkt/content-calendar",
"activities": [
"Check What Needs Clearing"
]
},
{
"process": "ref/mkt/develop-marketing-strategy",
"activities": [
"Check the Options against What Is Agreed",
"Issue the Strategy at a Version"
]
},
{
"process": "ref/mkt/engage-marketing-agency",
"activities": [
"Write the Scope",
"Brief the Agency"
]
},
{
"process": "ref/mkt/ideal-customer-profile",
"activities": [
"Issue the Profile as a Version"
]
},
{
"process": "ref/mkt/lead-scoring-model",
"activities": [
"Issue the Model as a Version"
]
},
{
"process": "ref/mkt/map-buyer-journey",
"activities": [
"Issue the Map as a Version"
]
},
{
"process": "ref/mkt/name-product",
"activities": [
"Issue the Name at a Version",
"Tell the Processes that Read It"
]
},
{
"process": "ref/mkt/plan-marketing-capacity",
"activities": [
"Check What an Abstract Agent Would Need"
]
},
{
"process": "ref/mkt/pricing-change",
"activities": [
"Set the Terms of the Change"
]
},
{
"process": "ref/mkt/prioritize-initiatives",
"activities": [
"Check the Requests against the Standards"
]
},
{
"process": "ref/mkt/product-positioning",
"activities": [
"Read the Standards in Force",
"Issue the Positioning at a Version"
]
},
{
"process": "ref/mkt/refresh-content",
"activities": [
"Score the Pages against the Current Standards"
]
},
{
"process": "ref/mkt/segment-the-market",
"activities": [
"Issue the Segments as a Version"
]
},
{
"process": "ref/mkt/select-marketing-technology",
"activities": [
"Check It against the Standards"
]
},
{
"process": "ref/mkt/size-market-opportunity",
"activities": [
"Issue the Sizing as a Version"
]
},
{
"process": "ref/mkt/verify-ad-placement",
"activities": [
"Confirm the Rules in Force",
"Update the Standing Lists"
]
}
]
}
}
}
README.mdWhat the package is, and why it carries no mcp.json.
# standards-keeper 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 eighteen 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/annual-marketing-plan`, `ref/mkt/approve-marketing-material`, `ref/mkt/attribution-model`, `ref/mkt/brand-audit`, `ref/mkt/brand-guidelines`, `ref/mkt/brand-health`, and 23 more. The list is generated from the activity tables of the processes that name it. https://agentcatalog.com/abstract-agents/standards-keeper
skills/brief-the-agents-that-work-to-the-standard/SKILL.mdRuns the session where the agents that write, draw or buy work through the standard on a real piece, so they hear the scope, the rules and the measures at the same time. Use it when running `ref/mkt/brand-audit` and 3 other reference processes.
--- name: brief-the-agents-that-work-to-the-standard description: Runs the session where the agents that write, draw or buy work through the standard on a real piece, so they hear the scope, the rules and the measures at the same time. Use it when running `ref/mkt/brand-audit` and 3 other reference processes. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Brief the agents that work to the standard ## What it does Runs the session where the agents that write, draw or buy work through the standard on a real piece, so they hear the scope, the rules and the measures at the same time. ## Where it happens The agent does this in four activities across four reference processes. Each one names the activity as that process words it. - **Brief the Agency** - `ref/mkt/engage-marketing-agency`, activity 15 - [Engage Marketing Agency](../../../../processes/marketing/engage-marketing-agency.md) - **Brief the Agents that Write to It** - `ref/mkt/brand-voice`, activity 17 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - **Brief the Audit** - `ref/mkt/brand-audit`, activity 4 - [Conduct Brand Audit](../../../../processes/marketing/brand-audit.md) - **Brief the Makers** - `ref/mkt/brand-guidelines`, activity 4 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) - `ref/mkt/brand-voice`, activity 4 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/check-a-proposal-against-what-is-agreed/SKILL.md"Reads something being proposed against the standards in force and says whether it fits: a strategic option, a request for work, a tool, a creative direction, or a set of planned topics that would need a claim, a permission or a legal read before it could run. Use it when running `ref/mkt/brand-identity` and 4 other reference processes."
--- name: check-a-proposal-against-what-is-agreed description: "Reads something being proposed against the standards in force and says whether it fits: a strategic option, a request for work, a tool, a creative direction, or a set of planned topics that would need a claim, a permission or a legal read before it could run. Use it when running `ref/mkt/brand-identity` and 4 other reference processes." license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Check a proposal against what is agreed ## What it does Reads something being proposed against the standards in force and says whether it fits: a strategic option, a request for work, a tool, a creative direction, or a set of planned topics that would need a claim, a permission or a legal read before it could run. ## Where it happens The agent does this in five activities across five reference processes. Each one names the activity as that process words it. - **Check It against the Standards** - `ref/mkt/select-marketing-technology`, activity 13 - [Select Marketing Technology](../../../../processes/marketing/select-marketing-technology.md) - **Check What Needs Clearing** - `ref/mkt/content-calendar`, activity 9 - [Plan Content Calendar](../../../../processes/marketing/content-calendar.md) - **Check the Options against What Is Agreed** - `ref/mkt/develop-marketing-strategy`, activity 9 - [Develop Marketing Strategy](../../../../processes/marketing/develop-marketing-strategy.md) - **Check the Requests against the Standards** - `ref/mkt/prioritize-initiatives`, activity 8 - [Prioritize Marketing Initiatives](../../../../processes/marketing/prioritize-initiatives.md) - **Test the Directions against the Positioning** - `ref/mkt/brand-identity`, activity 6 - [Develop Brand Identity](../../../../processes/marketing/brand-identity.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/choose-a-model/SKILL.mdChooses one model from the candidates and writes down the reasons it was chosen over the others. Use it when running `ref/mkt/attribution-model`.
--- name: choose-a-model description: Chooses one model from the candidates and writes down the reasons it was chosen over the others. Use it when running `ref/mkt/attribution-model`. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Choose a model ## What it does Chooses one model from the candidates and writes down the reasons it was chosen over the others. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Choose the Model** - `ref/mkt/attribution-model`, activity 10 - [Maintain Attribution Model](../../../../processes/marketing/attribution-model.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/collect-what-each-area-proposes/SKILL.mdAsks every area what it would run and why, in one format, and puts the answers into a single list. Use it when running `ref/mkt/annual-marketing-plan`.
--- name: collect-what-each-area-proposes description: Asks every area what it would run and why, in one format, and puts the answers into a single list. Use it when running `ref/mkt/annual-marketing-plan`. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Collect what each area proposes ## What it does Asks every area what it would run and why, in one format, and puts the answers into a single list. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Collect the Candidate Initiatives** - `ref/mkt/annual-marketing-plan`, activity 6 - [Build Annual Marketing Plan](../../../../processes/marketing/annual-marketing-plan.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/issue-a-standard-at-a-version/SKILL.mdTurns an agreed change into a numbered version with the date it applies from, so anything made after that date is judged against it and anything made earlier is still judged against the version that was current then. Use it when running `ref/mkt/attribution-model` and 14 other reference processes.
--- name: issue-a-standard-at-a-version description: Turns an agreed change into a numbered version with the date it applies from, so anything made after that date is judged against it and anything made earlier is still judged against the version that was current then. Use it when running `ref/mkt/attribution-model` and 14 other reference processes. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Issue a standard at a version ## What it does Turns an agreed change into a numbered version with the date it applies from, so anything made after that date is judged against it and anything made earlier is still judged against the version that was current then. ## Where it happens The agent does this in fourteen activities across fifteen reference processes. Each one names the activity as that process words it. - **Issue the Identity at a Version** - `ref/mkt/brand-refresh`, activity 2 - [Roll Out Brand Refresh](../../../../processes/marketing/brand-refresh.md) - **Issue the Map as a Version** - `ref/mkt/map-buyer-journey`, activity 16 - [Map Buyer Journey](../../../../processes/marketing/map-buyer-journey.md) - **Issue the Model as a Version** - `ref/mkt/lead-scoring-model`, activity 13 - [Maintain Lead Scoring Model](../../../../processes/marketing/lead-scoring-model.md) - **Issue the Model at a Version** - `ref/mkt/attribution-model`, activity 12 - [Maintain Attribution Model](../../../../processes/marketing/attribution-model.md) - **Issue the Name at a Version** - `ref/mkt/name-product`, activity 14 - [Name Product or Company](../../../../processes/marketing/name-product.md) - **Issue the Persona as a Version** - `ref/mkt/buyer-persona`, activity 17 - [Develop Buyer Persona](../../../../processes/marketing/buyer-persona.md) - **Issue the Positioning at a Version** - `ref/mkt/product-positioning`, activity 15 - [Develop Product Positioning](../../../../processes/marketing/product-positioning.md) - **Issue the Profile as a Version** - `ref/mkt/ideal-customer-profile`, activity 14 - [Define Ideal Customer Profile](../../../../processes/marketing/ideal-customer-profile.md) - **Issue the Segments as a Version** - `ref/mkt/segment-the-market`, activity 15 - [Segment the Market](../../../../processes/marketing/segment-the-market.md) - **Issue the Sizing as a Version** - `ref/mkt/size-market-opportunity`, activity 14 - [Size Market Opportunity](../../../../processes/marketing/size-market-opportunity.md) - **Issue the Standard at a Version** - `ref/mkt/brand-positioning`, activity 14 - [Define Brand Positioning](../../../../processes/marketing/brand-positioning.md) - `ref/mkt/brand-voice`, activity 15 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - **Issue the Strategy at a Version** - `ref/mkt/develop-marketing-strategy`, activity 14 - [Develop Marketing Strategy](../../../../processes/marketing/develop-marketing-strategy.md) - **Issue the Version** - `ref/mkt/brand-guidelines`, activity 12 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) - **Update the Standing Lists** - `ref/mkt/verify-ad-placement`, activity 13 - [Verify Ad Placement Quality](../../../../processes/marketing/verify-ad-placement.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/look-back-at-how-the-standard-is-working/SKILL.mdReports how the rules are behaving in practice, naming the ones that keep failing, the ones nobody applies, and what is worth repeating or avoiding the next time a version is written. Use it when running `ref/mkt/attribution-model` and 3 other reference processes.
--- name: look-back-at-how-the-standard-is-working description: Reports how the rules are behaving in practice, naming the ones that keep failing, the ones nobody applies, and what is worth repeating or avoiding the next time a version is written. Use it when running `ref/mkt/attribution-model` and 3 other reference processes. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Look back at how the standard is working ## What it does Reports how the rules are behaving in practice, naming the ones that keep failing, the ones nobody applies, and what is worth repeating or avoiding the next time a version is written. ## Where it happens The agent does this in two activities across four reference processes. Each one names the activity as that process words it. - **Record What Was Learned** - `ref/mkt/attribution-model`, activity 16 - [Maintain Attribution Model](../../../../processes/marketing/attribution-model.md) - `ref/mkt/brand-guidelines`, activity 16 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) - `ref/mkt/brand-refresh`, activity 26 - [Roll Out Brand Refresh](../../../../processes/marketing/brand-refresh.md) - **Watch How the Rules Score** - `ref/mkt/brand-voice`, activity 18 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/name-the-version-that-applies/SKILL.mdSays which standards apply to a piece of work and which version of each one is in force, and reads out the rules that bear on it, including any the proposed change would contradict. Use it when running `ref/mkt/approve-marketing-material` and 5 other reference processes.
--- name: name-the-version-that-applies description: Says which standards apply to a piece of work and which version of each one is in force, and reads out the rules that bear on it, including any the proposed change would contradict. Use it when running `ref/mkt/approve-marketing-material` and 5 other reference processes. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Name the version that applies ## What it does Says which standards apply to a piece of work and which version of each one is in force, and reads out the rules that bear on it, including any the proposed change would contradict. ## Where it happens The agent does this in six activities across six reference processes. Each one names the activity as that process words it. - **Confirm the Rules in Force** - `ref/mkt/verify-ad-placement`, activity 3 - [Verify Ad Placement Quality](../../../../processes/marketing/verify-ad-placement.md) - **Name the Standard Versions** - `ref/mkt/brand-audit`, activity 3 - [Conduct Brand Audit](../../../../processes/marketing/brand-audit.md) - **Name the Standards and the Signers** - `ref/mkt/approve-marketing-material`, activity 2 - [Approve Marketing Material](../../../../processes/marketing/approve-marketing-material.md) - **Read the Guidelines in Force** - `ref/mkt/brand-guidelines`, activity 2 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) - **Read the Promise the Brand Makes** - `ref/mkt/brand-health`, activity 2 - [Measure Brand Health](../../../../processes/marketing/brand-health.md) - **Read the Standards in Force** - `ref/mkt/product-positioning`, activity 2 - [Develop Product Positioning](../../../../processes/marketing/product-positioning.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/name-where-something-and-the-standard-do-not-agree/SKILL.mdSets what was found beside what the standard says and names each place the two do not agree, including where what people believe differs from what the organization claims. Use it when running `ref/mkt/brand-audit` and `ref/mkt/brand-health`.
--- name: name-where-something-and-the-standard-do-not-agree description: Sets what was found beside what the standard says and names each place the two do not agree, including where what people believe differs from what the organization claims. Use it when running `ref/mkt/brand-audit` and `ref/mkt/brand-health`. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Name where something and the standard do not agree ## What it does Sets what was found beside what the standard says and names each place the two do not agree, including where what people believe differs from what the organization claims. ## Where it happens The agent does this in two activities across two reference processes. Each one names the activity as that process words it. - **Name the Gaps** - `ref/mkt/brand-audit`, activity 13 - [Conduct Brand Audit](../../../../processes/marketing/brand-audit.md) - **Set Perception against the Promise** - `ref/mkt/brand-health`, activity 19 - [Measure Brand Health](../../../../processes/marketing/brand-health.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/release-a-version-and-retire-the-one-it-replaces/SKILL.mdPuts an issued version where the people who work to it can get it, with the page live and the files released, and takes the version it replaced out of circulation. Use it when running `ref/mkt/brand-guidelines`, `ref/mkt/brand-identity` and `ref/mkt/brand-refresh`.
--- name: release-a-version-and-retire-the-one-it-replaces description: Puts an issued version where the people who work to it can get it, with the page live and the files released, and takes the version it replaced out of circulation. Use it when running `ref/mkt/brand-guidelines`, `ref/mkt/brand-identity` and `ref/mkt/brand-refresh`. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Release a version and retire the one it replaces ## What it does Puts an issued version where the people who work to it can get it, with the page live and the files released, and takes the version it replaced out of circulation. ## Where it happens The agent does this in three activities across three reference processes. Each one names the activity as that process words it. - **Issue the Rules and Release the Files** - `ref/mkt/brand-identity`, activity 15 - [Develop Brand Identity](../../../../processes/marketing/brand-identity.md) - **Publish the Version** - `ref/mkt/brand-guidelines`, activity 13 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) - **Retire the Old Identity** - `ref/mkt/brand-refresh`, activity 23 - [Roll Out Brand Refresh](../../../../processes/marketing/brand-refresh.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/say-what-an-agent-would-need-to-hold-the-work/SKILL.md"Says what an abstract agent would have to be given before it could hold a piece of work: the grants it needs, the standards it works to, and what it has to put to a person. Use it when running `ref/mkt/plan-marketing-capacity`."
--- name: say-what-an-agent-would-need-to-hold-the-work description: "Says what an abstract agent would have to be given before it could hold a piece of work: the grants it needs, the standards it works to, and what it has to put to a person. Use it when running `ref/mkt/plan-marketing-capacity`." license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Say what an agent would need to hold the work ## What it does Says what an abstract agent would have to be given before it could hold a piece of work: the grants it needs, the standards it works to, and what it has to put to a person. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Check What an Abstract Agent Would Need** - `ref/mkt/plan-marketing-capacity`, activity 8 - [Plan Marketing Capacity](../../../../processes/marketing/plan-marketing-capacity.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/score-material-against-the-standard/SKILL.mdScores each item in a sample of material against the rule it has to meet, at the version of that rule which was in force, and records the verdict with the rule behind it. Use it when running `ref/mkt/brand-audit` and `ref/mkt/refresh-content`.
--- name: score-material-against-the-standard description: Scores each item in a sample of material against the rule it has to meet, at the version of that rule which was in force, and records the verdict with the rule behind it. Use it when running `ref/mkt/brand-audit` and `ref/mkt/refresh-content`. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Score material against the standard ## What it does Scores each item in a sample of material against the rule it has to meet, at the version of that rule which was in force, and records the verdict with the rule behind it. ## Where it happens The agent does this in two activities across two reference processes. Each one names the activity as that process words it. - **Score the Pages against the Current Standards** - `ref/mkt/refresh-content`, activity 5 - [Refresh Published Content](../../../../processes/marketing/refresh-content.md) - **Score the Sample** - `ref/mkt/brand-audit`, activity 7 - [Conduct Brand Audit](../../../../processes/marketing/brand-audit.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/send-a-finding-to-the-agent-that-can-fix-it/SKILL.mdSends each finding to the agent that can close it, and sends a rule or a claim that no longer holds back to be written again. Use it when running `ref/mkt/brand-audit` and `ref/mkt/competitive-battlecard`.
--- name: send-a-finding-to-the-agent-that-can-fix-it description: Sends each finding to the agent that can close it, and sends a rule or a claim that no longer holds back to be written again. Use it when running `ref/mkt/brand-audit` and `ref/mkt/competitive-battlecard`. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Send a finding to the agent that can fix it ## What it does Sends each finding to the agent that can close it, and sends a rule or a claim that no longer holds back to be written again. ## Where it happens The agent does this in two activities across two reference processes. Each one names the activity as that process words it. - **Hand on the Claim Change** - `ref/mkt/competitive-battlecard`, activity 15 - [Refresh Competitive Battlecard](../../../../processes/marketing/competitive-battlecard.md) - **Hand the Gaps to the Owners** - `ref/mkt/brand-audit`, activity 16 - [Conduct Brand Audit](../../../../processes/marketing/brand-audit.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/set-the-scope-and-the-dates-of-the-work/SKILL.mdWrites down what a piece of work covers, who owes which part of it and by when, and the conditions under which it gets looked at again. Use it when running `ref/mkt/attribution-model`, `ref/mkt/brand-guidelines` and `ref/mkt/engage-marketing-agency`.
--- name: set-the-scope-and-the-dates-of-the-work description: Writes down what a piece of work covers, who owes which part of it and by when, and the conditions under which it gets looked at again. Use it when running `ref/mkt/attribution-model`, `ref/mkt/brand-guidelines` and `ref/mkt/engage-marketing-agency`. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Set the scope and the dates of the work ## What it does Writes down what a piece of work covers, who owes which part of it and by when, and the conditions under which it gets looked at again. ## Where it happens The agent does this in four activities across three reference processes. Each one names the activity as that process words it. - **Plan the Version** - `ref/mkt/brand-guidelines`, activity 3 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) - **Set What the Review Covers** - `ref/mkt/attribution-model`, activity 3 - [Maintain Attribution Model](../../../../processes/marketing/attribution-model.md) - **Set the Review Conditions** - `ref/mkt/attribution-model`, activity 15 - [Maintain Attribution Model](../../../../processes/marketing/attribution-model.md) - **Write the Scope** - `ref/mkt/engage-marketing-agency`, activity 4 - [Engage Marketing Agency](../../../../processes/marketing/engage-marketing-agency.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/set-the-terms-of-a-change/SKILL.md"Settles the terms a change goes out under: who stays on the old arrangement, how much notice people get, and what the new one is. Use it when running `ref/mkt/pricing-change`."
--- name: set-the-terms-of-a-change description: "Settles the terms a change goes out under: who stays on the old arrangement, how much notice people get, and what the new one is. Use it when running `ref/mkt/pricing-change`." license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Set the terms of a change ## What it does Settles the terms a change goes out under: who stays on the old arrangement, how much notice people get, and what the new one is. ## 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 Terms of the Change** - `ref/mkt/pricing-change`, activity 3 - [Launch Pricing Change](../../../../processes/marketing/pricing-change.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/settle-what-a-rule-means/SKILL.mdAnswers a scoring agent that cannot tell from the text what a rule requires, and writes the answer down against the rule so everyone reading it afterwards gets the same answer. Use it when running `ref/mkt/approve-marketing-material`.
--- name: settle-what-a-rule-means description: Answers a scoring agent that cannot tell from the text what a rule requires, and writes the answer down against the rule so everyone reading it afterwards gets the same answer. Use it when running `ref/mkt/approve-marketing-material`. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Settle what a rule means ## What it does Answers a scoring agent that cannot tell from the text what a rule requires, and writes the answer down against the rule so everyone reading it afterwards gets the same answer. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Settle a Disputed Rule** - `ref/mkt/approve-marketing-material`, activity 8 - [Approve Marketing Material](../../../../processes/marketing/approve-marketing-material.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/take-in-what-the-standard-has-to-carry/SKILL.mdTakes in an approved decision or a trigger and says what the standard now has to cover, which existing rules it touches, and by when the work has to be done. Use it when running `ref/mkt/attribution-model` and 4 other reference processes.
--- name: take-in-what-the-standard-has-to-carry description: Takes in an approved decision or a trigger and says what the standard now has to cover, which existing rules it touches, and by when the work has to be done. Use it when running `ref/mkt/attribution-model` and 4 other reference processes. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Take in what the standard has to carry ## What it does Takes in an approved decision or a trigger and says what the standard now has to cover, which existing rules it touches, and by when the work has to be done. ## Where it happens The agent does this in five activities across five reference processes. Each one names the activity as that process words it. - **Take in the Approved Identity** - `ref/mkt/brand-refresh`, activity 1 - [Roll Out Brand Refresh](../../../../processes/marketing/brand-refresh.md) - **Take in the Approved Positioning** - `ref/mkt/brand-voice`, activity 1 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - **Take in the Change** - `ref/mkt/brand-guidelines`, activity 1 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) - **Take in the Positioning** - `ref/mkt/brand-identity`, activity 1 - [Develop Brand Identity](../../../../processes/marketing/brand-identity.md) - **Take in the Trigger** - `ref/mkt/attribution-model`, activity 1 - [Maintain Attribution Model](../../../../processes/marketing/attribution-model.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/tell-the-agents-that-read-a-standard-what-changed/SKILL.mdTells every agent and process that reads a standard that a new version is in force, what changed in it, and from when, and records that each one confirmed it has the new version. Use it when running `ref/mkt/attribution-model` and 6 other reference processes.
--- name: tell-the-agents-that-read-a-standard-what-changed description: Tells every agent and process that reads a standard that a new version is in force, what changed in it, and from when, and records that each one confirmed it has the new version. Use it when running `ref/mkt/attribution-model` and 6 other reference processes. license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Tell the agents that read a standard what changed ## What it does Tells every agent and process that reads a standard that a new version is in force, what changed in it, and from when, and records that each one confirmed it has the new version. ## Where it happens The agent does this in four activities across seven reference processes. Each one names the activity as that process words it. - **Hand the Decisions On** - `ref/mkt/brief-executive-leadership`, activity 19 - [Brief Executive Leadership](../../../../processes/marketing/brief-executive-leadership.md) - **Tell Everyone Who Works to the Brand** - `ref/mkt/brand-guidelines`, activity 14 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) - **Tell the Processes that Read It** - `ref/mkt/brand-positioning`, activity 15 - [Define Brand Positioning](../../../../processes/marketing/brand-positioning.md) - `ref/mkt/brand-refresh`, activity 3 - [Roll Out Brand Refresh](../../../../processes/marketing/brand-refresh.md) - `ref/mkt/brand-voice`, activity 16 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - `ref/mkt/name-product`, activity 15 - [Name Product or Company](../../../../processes/marketing/name-product.md) - **Tell the Processes that Read the Model** - `ref/mkt/attribution-model`, activity 14 - [Maintain Attribution Model](../../../../processes/marketing/attribution-model.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
skills/write-rules-that-can-be-scored/SKILL.md"Turns agreed intent into rules another agent can apply without asking anyone: each rule numbered and written against something checkable, with the words, the sizes, the values and the examples that separate a pass from a failure. Use it when running `ref/mkt/brand-guidelines` and 3 other reference processes."
--- name: write-rules-that-can-be-scored description: "Turns agreed intent into rules another agent can apply without asking anyone: each rule numbered and written against something checkable, with the words, the sizes, the values and the examples that separate a pass from a failure. Use it when running `ref/mkt/brand-guidelines` and 3 other reference processes." license: CC-BY-4.0 metadata: agent: standards-keeper agent-version: "1" --- # Write rules that can be scored ## What it does Turns agreed intent into rules another agent can apply without asking anyone: each rule numbered and written against something checkable, with the words, the sizes, the values and the examples that separate a pass from a failure. ## Where it happens The agent does this in five activities across four reference processes. Each one names the activity as that process words it. - **Set the Voice Rules** - `ref/mkt/brand-positioning`, activity 12 - [Define Brand Positioning](../../../../processes/marketing/brand-positioning.md) - **Turn the Voice into Scorable Rules** - `ref/mkt/brand-voice`, activity 11 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - **Write the Rules** - `ref/mkt/brand-guidelines`, activity 5 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) - **Write the Usage Rules** - `ref/mkt/brand-identity`, activity 13 - [Develop Brand Identity](../../../../processes/marketing/brand-identity.md) - **Write the Word List** - `ref/mkt/brand-voice`, activity 10 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) ## What to record Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change. 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`.
mcp.jsonNot in the package.
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.
Setting 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 5 roles, and the access level beside each one is the most the work needs.
standards-store · writeinterpretation-record · writenotification-channel · writeapproval-record · readprocess-bindings · read
mcp.json
You write this file. It is not part of the package.
{
"$schema": "https://agent-plugins.org/schemas/1.1.0/mcp.schema.json",
"mcpServers": {
"standards-store": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/standards-store"
},
"interpretation-record": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/interpretation-record"
},
"notification-channel": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/notification-channel"
},
"approval-record": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/approval-record"
},
"process-bindings": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/process-bindings"
}
}
}
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.
Where it works
50 processes, 122 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.
Build Annual Marketing Planref/mkt/annual-marketing-plan · 16 activities
Collect the Candidate Initiatives
Approve Marketing Materialref/mkt/approve-marketing-material · 13 activities
Name the Standards and the Signers · Settle a Disputed Rule
Maintain Attribution Modelref/mkt/attribution-model · 16 activities
Take in the Trigger · Set What the Review Covers · Choose the Model · Issue the Model at a Version · Tell the Processes that Read the Model · Set the Review Conditions · Record What Was Learned
Conduct Brand Auditref/mkt/brand-audit · 17 activities
Name the Standard Versions · Brief the Audit · Score the Sample · Name the Gaps · Hand the Gaps to the Owners
Publish Brand Guidelinesref/mkt/brand-guidelines · 16 activities
Take in the Change · Read the Guidelines in Force · Plan the Version · Brief the Makers · Write the Rules · Issue the Version · Publish the Version · Tell Everyone Who Works to the Brand · Record What Was Learned
Measure Brand Healthref/mkt/brand-health · 23 activities
Read the Promise the Brand Makes · Set Perception against the Promise
Develop Brand Identityref/mkt/brand-identity · 16 activities
Take in the Positioning · Test the Directions against the Positioning · Write the Usage Rules · Issue the Rules and Release the Files
Define Brand Positioningref/mkt/brand-positioning · 16 activities
Set the Voice Rules · Issue the Standard at a Version · Tell the Processes that Read It
Roll Out Brand Refreshref/mkt/brand-refresh · 26 activities
Take in the Approved Identity · Issue the Identity at a Version · Tell the Processes that Read It · Retire the Old Identity · Record What Was Learned
Define Brand Voice and Messagingref/mkt/brand-voice · 19 activities
Take in the Approved Positioning · Brief the Makers · Write the Word List · Turn the Voice into Scorable Rules · Issue the Standard at a Version · Tell the Processes that Read It · Brief the Agents that Write to It · Watch How the Rules Score
Brief Executive Leadershipref/mkt/brief-executive-leadership · 20 activities
Hand the Decisions On
Develop Buyer Personaref/mkt/buyer-persona · 20 activities
Issue the Persona as a Version
Refresh Competitive Battlecardref/mkt/competitive-battlecard · 16 activities
Hand on the Claim Change
Plan Content Calendarref/mkt/content-calendar · 15 activities
Check What Needs Clearing
Develop Marketing Strategyref/mkt/develop-marketing-strategy · 17 activities
Check the Options against What Is Agreed · Issue the Strategy at a Version
Engage Marketing Agencyref/mkt/engage-marketing-agency · 16 activities
Write the Scope · Brief the Agency
Define Ideal Customer Profileref/mkt/ideal-customer-profile · 17 activities
Issue the Profile as a Version
Maintain Lead Scoring Modelref/mkt/lead-scoring-model · 18 activities
Issue the Model as a Version
Map Buyer Journeyref/mkt/map-buyer-journey · 19 activities
Issue the Map as a Version
Name Product or Companyref/mkt/name-product · 16 activities
Issue the Name at a Version · Tell the Processes that Read It
Plan Marketing Capacityref/mkt/plan-marketing-capacity · 15 activities
Check What an Abstract Agent Would Need
Launch Pricing Changeref/mkt/pricing-change · 16 activities
Set the Terms of the Change
Prioritize Marketing Initiativesref/mkt/prioritize-initiatives · 16 activities
Check the Requests against the Standards
Develop Product Positioningref/mkt/product-positioning · 17 activities
Read the Standards in Force · Issue the Positioning at a Version
Refresh Published Contentref/mkt/refresh-content · 22 activities
Score the Pages against the Current Standards
Segment the Marketref/mkt/segment-the-market · 18 activities
Issue the Segments as a Version
Select Marketing Technologyref/mkt/select-marketing-technology · 21 activities
Check It against the Standards
Size Market Opportunityref/mkt/size-market-opportunity · 16 activities
Issue the Sizing as a Version
Verify Ad Placement Qualityref/mkt/verify-ad-placement · 15 activities
Confirm the Rules in Force · Update the Standing Lists
Administer a Toolref/rev/administer-a-tool · 15 activities
Hold the Configuration at a Version
Answer a Data Questionref/rev/answer-a-data-question · 14 activities
Name the Measures and Versions · Check It Against the Definitions
Approve a Discountref/rev/approve-a-discount · 15 activities
Find the Level That Applies
Build a Reportref/rev/build-a-report · 14 activities
Settle the Measures It Uses
Certify a Sellerref/rev/certify-a-seller · 17 activities
Name What Must Be Proved · Set the Standard for a Pass · Check the Assessment Is Fair · Record the Certification
Choose a Toolref/rev/choose-a-tool · 18 activities
Brief the Evaluation · Check It against the Standards
Define a Measureref/rev/define-a-measure · 16 activities
Read the Definitions in Force · Write the Rule for What Counts · Issue the Definition at a Version · Tell the Reports that Read It · Set the Review Condition
Document a Processref/rev/document-a-process · 16 activities
Check It Against the Standards · Set the Version and Publish
Instrument a Processref/rev/instrument-a-process · 16 activities
Check the Watching Against Policy
Integrate a Toolref/rev/integrate-a-tool · 16 activities
Agree What Flows Where · Check the Definitions Match · Write down What Flows Where
Onboard a Sellerref/rev/onboard-a-seller · 17 activities
Teach the Process
Publish the Approval Thresholdsref/rev/publish-the-approval-thresholds · 16 activities
Take in the Reason to Publish · Find Where the Table Fails Today · Draft the Levels · Set the Time Promised per Level · Name a Delegate for Each Signer · Check the Levels Against Policy · Get Each Signer to Accept · Version the Table · Publish Where Sellers Work · Tell the Sellers What Changed · Review the Table on a Date
Retire a Reportref/rev/retire-a-report · 16 activities
Read the Retirement Conditions · Keep the Record
Retire a Toolref/rev/retire-a-tool · 15 activities
Say What Data Is Kept · Check the Export Reads · Review before Switch-Off
Review Adoptionref/rev/review-adoption · 16 activities
Read the Exceptions Already Granted · Open the Enforcement Changes
Review a Non-Standard Dealref/rev/review-a-non-standard-deal · 16 activities
Feed the Pattern to the Standards
Roll Out a Changeref/rev/roll-out-a-change · 17 activities
Change the Process and the Rules · Watch It in Live Deals
Answer a Security Questionnaireref/sls/answer-a-security-questionnaire · 15 activities
Match Against the Library · Check the Matches Are Still True · Check What Must Not Be Said · Feed the Library
Configure and Priceref/sls/configure-and-price · 14 activities
Price Against the Rate Card
Respond to an RFPref/sls/respond-to-an-rfp · 17 activities
Answer the Security and Legal Parts
Write a Proposalref/sls/write-a-proposal · 15 activities
Pull the Reusable Parts · Check the Claims
The records it leaves
What it writes down
Every standard at a numbered version, with the date it took effect and the date it stopped being current. A change record for each version, saying what changed, what it replaced, who approved it, and when. Every earlier version left readable at its own number, so work can be judged against the standard it was made under. Every interpretation it issued, attached to the rule and the version it interprets, carrying the question that prompted it and the agent that asked. A dated record of who was told about each change.
Where it stops
What it will not do
Does not score anything against the standards it holds. That is the brand-check agent, reading the claims register and the brand guidelines at their versions, and this agent's job is to answer that agent when a rule is unclear. Does not write the standards: what the positioning and the promise say is settled by the people who decide, the decision-coordinator agent runs that decision, and the agreed answer comes here to be versioned. Does not edit a piece against the voice rules, which is the editor agent working on structure and argument and the copywriter agent writing the words. Does not add a claim to the claims register on its own, and does not approve one. Does not decide what the organization publishes or when, which is the content-manager agent for one piece and the content-librarian agent for everything already live. Does not hold the record of who has agreed to be contacted, which is the consent-manager agent. 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.
Agents built from this plugin
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.
declare it
implements: standards-keeper v1
One line in your agent's manifest. The catalog checks it against the records your runs actually leave.
A guess, not a claim
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.
Might do the job
Things that could be the worker: an agent, a packaged product, a plugin.
- The Website Specification — A2A agent Agent · specification.website Read-only A2A agent that answers natural-language questions about The Website Specification and returns matching topics with status, canonical URL,…
Might be reached for
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.
- mcp-server MCP server · iris-eval.github.io · run it yourself · mirrored from the MCP registry The agent eval standard for MCP. Score every agent output for quality, safety, and cost.
- Pre-Trip compliance scanner MCP server · kesey.agency · run it yourself · mirrored from the MCP registry Screen regulated-health marketing copy against source-cited rulesets, all 50 states.
- kage MCP server · kage-core.com · run it yourself · mirrored from the MCP registry Verified memory for coding agents: claims cited against code, stale withheld, savings receipts.