Certification check
certification-check v1Tests that somebody can actually do the parts of a job a customer would notice, and says no when they cannot, before a customer is the one finding out. Names what must be proved, checks the assessment tests the standard it claims to, picks assessors who are not judging their own people, runs the sitting, and scores every part against a written pass line with the evidence kept next to the score. It carries the result to a person and never grants a pass itself. It records the standard version and the expiry, and it reports afterwards whether the people it passed went on to do the job, because a standard that passes people who cannot sell is not a gate.
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 assessment for every part it tests: the fairness check on the assessment itself, the knowledge check, and the short check that a change was understood. Convenes the briefing where every assessor hears the same standard, the same scoring sheet and the same worked examples at once. Convenes the approval where a named person signs a pass, a conditional pass or a retake, and it holds that room open rather than deciding in it. Runs roll-call on attendance against the list of who must be told, and collect-and-report on the sitting as it runs and on what certification predicted months later. Takes the standard from the standards-keeper agent and the assessment from the curriculum-manager agent, hands failures back to that agent as named gaps, and hands the signed pass to the onboarding-manager agent before any customer conversation is booked.
The plugin
What is in the package
8 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.
certification-check · 13 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 teach, and does not write the courses or the questions, which is the curriculum-manager agent's work. Does not write the standard it scores against, which belongs to the standards-keeper agent, and an assessment that tests something the standard does not name is changed rather than scored. Does not decide who is put forward or run the ramp around them, which is the onboarding-manager agent's job. Does not lower a pass mark, move it after a sitting, or issue a certification on its own authority. ## What always goes to a person - a pass nobody will sign, to the named signer, with the run held. This agent never grants a certification itself - an assessor scoring somebody they manage or hired, to the sales manager, with the sitting stopped at that part and the first score kept in the record marked as set aside - assessors who cannot agree after rereading the evidence, to the named decider, and the disagreement onward to the standards-keeper agent as a case the standard did not cover - a certificate asked for on a practical nobody watched, refused, and the request goes to whoever asked for it. Speed is not a reason recorded anywhere - a standard that did not predict results, to whoever owns the standard, with the report attached, so it is amended at a version rather than quietly reused - a certificate approaching expiry or a certified person who has left, to whoever owns the relationship, before a deal in flight discovers it
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 `certification-check` leaves
the following behind, on every run.
Per person: the request, naming who is put forward, for which job and by
when. The parts that must be proved, each with the reason a customer
would notice it. The assessment as checked, item by item, each item
naming the part of the standard it tests, and every item removed with
the reason. The assessors, and the conflicts that were ruled out. What
the assessors were briefed on, at a version. The sitting as scheduled,
including what the person was told they may bring. A score per part,
with the evidence it was given on and the assessor who gave it, kept so
a disputed score can be reread rather than re-argued. A score set aside
for conflict is kept marked as set aside, never deleted. The signed
result, who signed it, what it covers, and the condition and expiry on
any conditional pass. The certificate with its issue date, requirement
version and expiry. Attendance by name against the list of who must be
told. The later report setting scores against what those people did.
An agent that leaves less than this does not implement
`certification-check`, 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
Seven 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 certification record | write | The result, the standard version, the assessors, the evidence and the expiry date. |
| the learning system | write | Who sat what, and who is due to sit again. |
| the enablement library | read | The material people were taught from, so the sitting tests what was covered. |
| the CRM | write | On a practice recor, the systems check is done on a practice record and never on a live customer. |
| the quoting system | write | On a practice recor, the same, for the quote the person builds unaided. |
| the partner portal | write | Certificates issued with a date and an expiry, and the certified list visible to both sides. |
| analytics | read | What the people who passed went on to do. |
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": "certification-check",
"version": "1.0.0",
"description": "The certification-check abstract agent. Tests that somebody can actually do the parts of a job a customer would notice, and says no when they cannot, before a customer is the one finding out.",
"author": {
"name": "AgentCatalog",
"url": "https://agentcatalog.com"
},
"homepage": "https://agentcatalog.com/abstract-agents/certification-check",
"repository": "https://github.com/jeffrschneider/agentcatalog",
"license": "CC-BY-4.0",
"keywords": [
"abstract-agent",
"reference",
"revenue-operations",
"certify-a-seller",
"certify-a-partner",
"roll-out-a-change",
"abstract"
],
"extensions": {
"com.agentcatalog.agent": {
"agent": "certification-check",
"agentVersion": 1,
"department": "revenue-operations",
"abstract": true,
"works": "Runs assessment for every part it tests: the fairness check on the assessment itself, the knowledge check, and the short check that a change was understood. Convenes the briefing where every assessor hears the same standard, the same scoring sheet and the same worked examples at once. Convenes the approval where a named person signs a pass, a conditional pass or a retake, and it holds that room open rather than deciding in it. Runs roll-call on attendance against the list of who must be told, and collect-and-report on the sitting as it runs and on what certification predicted months later. Takes the standard from the standards-keeper agent and the assessment from the curriculum-manager agent, hands failures back to that agent as named gaps, and hands the signed pass to the onboarding-manager agent before any customer conversation is booked.",
"servers": [
{
"name": "the certification record",
"role": "certification-record",
"access": "write",
"needs": "The result, the standard version, the assessors, the evidence and the expiry date."
},
{
"name": "the learning system",
"role": "learning-system",
"access": "write",
"needs": "Who sat what, and who is due to sit again."
},
{
"name": "the enablement library",
"role": "enablement-library",
"access": "read",
"needs": "The material people were taught from, so the sitting tests what was covered."
},
{
"name": "the CRM",
"role": "crm",
"access": "write",
"needs": "On a practice recor, the systems check is done on a practice record and never on a live customer."
},
{
"name": "the quoting system",
"role": "quoting-system",
"access": "write",
"needs": "On a practice recor, the same, for the quote the person builds unaided."
},
{
"name": "the partner portal",
"role": "partner-portal",
"access": "write",
"needs": "Certificates issued with a date and an expiry, and the certified list visible to both sides."
},
{
"name": "analytics",
"role": "analytics",
"access": "read",
"needs": "What the people who passed went on to do."
}
],
"records": [
"Per person: the request, naming who is put forward, for which job and by when.",
"The parts that must be proved, each with the reason a customer would notice it.",
"The assessment as checked, item by item, each item naming the part of the standard it tests, and every item removed with the reason.",
"The assessors, and the conflicts that were ruled out.",
"What the assessors were briefed on, at a version.",
"The sitting as scheduled, including what the person was told they may bring.",
"A score per part, with the evidence it was given on and the assessor who gave it, kept so a disputed score can be reread rather than re-argued.",
"A score set aside for conflict is kept marked as set aside, never deleted.",
"The signed result, who signed it, what it covers, and the condition and expiry on any conditional pass.",
"The certificate with its issue date, requirement version and expiry.",
"Attendance by name against the list of who must be told.",
"The later report setting scores against what those people did."
],
"escalates": [
"a pass nobody will sign, to the named signer, with the run held. This agent never grants a certification itself",
"an assessor scoring somebody they manage or hired, to the sales manager, with the sitting stopped at that part and the first score kept in the record marked as set aside",
"assessors who cannot agree after rereading the evidence, to the named decider, and the disagreement onward to the standards-keeper agent as a case the standard did not cover",
"a certificate asked for on a practical nobody watched, refused, and the request goes to whoever asked for it. Speed is not a reason recorded anywhere",
"a standard that did not predict results, to whoever owns the standard, with the report attached, so it is amended at a version rather than quietly reused",
"a certificate approaching expiry or a certified person who has left, to whoever owns the relationship, before a deal in flight discovers it"
],
"not": [
"Does not teach, and does not write the courses or the questions, which is the curriculum-manager agent's work.",
"Does not write the standard it scores against, which belongs to the standards-keeper agent, and an assessment that tests something the standard does not name is changed rather than scored.",
"Does not decide who is put forward or run the ramp around them, which is the onboarding-manager agent's job.",
"Does not lower a pass mark, move it after a sitting, or issue a certification on its own authority."
],
"processes": [
{
"process": "ref/prt/certify-a-partner",
"activities": [
"Sit the Written Exam",
"Assess the Practical",
"Decide Pass or Fail",
"Issue the Certificates"
]
},
{
"process": "ref/prt/run-a-first-deal-together",
"activities": [
"Decide They Can Do It Unaided"
]
},
{
"process": "ref/rev/certify-a-seller",
"activities": [
"Take in the Request to Certify",
"Name What Must Be Proved",
"Check the Assessment Is Fair",
"Name the Assessors",
"Brief the Assessors",
"Schedule the Sitting",
"Run the Knowledge Check",
"Score Against the Standard",
"Decide Pass or Retake",
"Record the Certification",
"Report What Certification Predicted"
]
},
{
"process": "ref/rev/onboard-a-seller",
"activities": [
"Prove the Parts that Matter"
]
},
{
"process": "ref/rev/roll-out-a-change",
"activities": [
"Check Who Attended",
"Check That It Landed"
]
}
]
}
}
}
README.mdWhat the package is, and why it carries no mcp.json.
# certification-check 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 eight 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/prt/certify-a-partner`, `ref/prt/run-a-first-deal-together`, `ref/rev/certify-a-seller`, `ref/rev/onboard-a-seller`, `ref/rev/roll-out-a-change`. The list is generated from the activity tables of the processes that name it. https://agentcatalog.com/abstract-agents/certification-check
skills/carry-the-result-to-a-person-who-signs-it/SKILL.mdPuts the result in front of the named signer as a pass, a conditional pass, a retake or a fail, and holds that room open rather than deciding in it. Use it when running `ref/prt/certify-a-partner`, `ref/prt/run-a-first-deal-together` and `ref/rev/certify-a-seller`.
--- name: carry-the-result-to-a-person-who-signs-it description: Puts the result in front of the named signer as a pass, a conditional pass, a retake or a fail, and holds that room open rather than deciding in it. Use it when running `ref/prt/certify-a-partner`, `ref/prt/run-a-first-deal-together` and `ref/rev/certify-a-seller`. license: CC-BY-4.0 metadata: agent: certification-check agent-version: "1" --- # Carry the result to a person who signs it ## What it does Puts the result in front of the named signer as a pass, a conditional pass, a retake or a fail, and holds that room open rather than deciding in it. The same signature is what says a partner needs no escort on the next deal. This agent never grants a pass itself. ## Where it happens The agent does this in three activities across three reference processes. Each one names the activity as that process words it. - **Decide Pass or Fail** - `ref/prt/certify-a-partner`, activity 12 - [Certify a Partner](../../../../processes/partners/certify-a-partner.md) - **Decide Pass or Retake** - `ref/rev/certify-a-seller`, activity 14 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) - **Decide They Can Do It Unaided** - `ref/prt/run-a-first-deal-together`, activity 15 - [Run a First Deal Together](../../../../processes/partners/run-a-first-deal-together.md) ## What to record Per person: the request, naming who is put forward, for which job and by when. The parts that must be proved, each with the reason a customer would notice it. The assessment as checked, item by item, each item naming the part of the standard it tests, and every item removed with the reason. The assessors, and the conflicts that were ruled out. What the assessors were briefed on, at a version. The sitting as scheduled, including what the person was told they may bring. A score per part, with the evidence it was given on and the assessor who gave it, kept so a disputed score can be reread rather than re-argued. A score set aside for conflict is kept marked as set aside, never deleted. The signed result, who signed it, what it covers, and the condition and expiry on any conditional pass. The certificate with its issue date, requirement version and expiry. Attendance by name against the list of who must be told. The later report setting scores against what those people did. 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-change-reached-the-people-it-names/SKILL.mdCalls the roll against the list of who must be told, and runs the short check that the change is actually understood by the people who sat through it. Use it when running `ref/rev/roll-out-a-change`.
--- name: check-a-change-reached-the-people-it-names description: Calls the roll against the list of who must be told, and runs the short check that the change is actually understood by the people who sat through it. Use it when running `ref/rev/roll-out-a-change`. license: CC-BY-4.0 metadata: agent: certification-check agent-version: "1" --- # Check a change reached the people it names ## What it does Calls the roll against the list of who must be told, and runs the short check that the change is actually understood by the people who sat through it. ## Where it happens The agent does this in two activities across one reference process. Each one names the activity as that process words it. - **Check That It Landed** - `ref/rev/roll-out-a-change`, activity 14 - [Roll Out a Change](../../../../processes/revenue-operations/roll-out-a-change.md) - **Check Who Attended** - `ref/rev/roll-out-a-change`, activity 13 - [Roll Out a Change](../../../../processes/revenue-operations/roll-out-a-change.md) ## What to record Per person: the request, naming who is put forward, for which job and by when. The parts that must be proved, each with the reason a customer would notice it. The assessment as checked, item by item, each item naming the part of the standard it tests, and every item removed with the reason. The assessors, and the conflicts that were ruled out. What the assessors were briefed on, at a version. The sitting as scheduled, including what the person was told they may bring. A score per part, with the evidence it was given on and the assessor who gave it, kept so a disputed score can be reread rather than re-argued. A score set aside for conflict is kept marked as set aside, never deleted. The signed result, who signed it, what it covers, and the condition and expiry on any conditional pass. The certificate with its issue date, requirement version and expiry. Attendance by name against the list of who must be told. The later report setting scores against what those people did. 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/put-a-sitting-in-front-of-fair-assessors/SKILL.mdPicks assessors who are not scoring their own people, briefs all of them on the same standard, the same scoring sheet and the same worked examples at once, and sets the date, the order and what the person may bring in with them. Use it when running `ref/rev/certify-a-seller`.
--- name: put-a-sitting-in-front-of-fair-assessors description: Picks assessors who are not scoring their own people, briefs all of them on the same standard, the same scoring sheet and the same worked examples at once, and sets the date, the order and what the person may bring in with them. Use it when running `ref/rev/certify-a-seller`. license: CC-BY-4.0 metadata: agent: certification-check agent-version: "1" --- # Put a sitting in front of fair assessors ## What it does Picks assessors who are not scoring their own people, briefs all of them on the same standard, the same scoring sheet and the same worked examples at once, and sets the date, the order and what the person may bring in with them. ## Where it happens The agent does this in three activities across one reference process. Each one names the activity as that process words it. - **Brief the Assessors** - `ref/rev/certify-a-seller`, activity 7 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) - **Name the Assessors** - `ref/rev/certify-a-seller`, activity 6 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) - **Schedule the Sitting** - `ref/rev/certify-a-seller`, activity 8 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) ## What to record Per person: the request, naming who is put forward, for which job and by when. The parts that must be proved, each with the reason a customer would notice it. The assessment as checked, item by item, each item naming the part of the standard it tests, and every item removed with the reason. The assessors, and the conflicts that were ruled out. What the assessors were briefed on, at a version. The sitting as scheduled, including what the person was told they may bring. A score per part, with the evidence it was given on and the assessor who gave it, kept so a disputed score can be reread rather than re-argued. A score set aside for conflict is kept marked as set aside, never deleted. The signed result, who signed it, what it covers, and the condition and expiry on any conditional pass. The certificate with its issue date, requirement version and expiry. Attendance by name against the list of who must be told. The later report setting scores against what those people did. 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/record-and-issue-what-was-passed/SKILL.mdRecords the result with the standard version, the assessors, the evidence and the expiry date, and issues the certificate with its issue date and its expiry where both sides can see it. Use it when running `ref/prt/certify-a-partner` and `ref/rev/certify-a-seller`.
--- name: record-and-issue-what-was-passed description: Records the result with the standard version, the assessors, the evidence and the expiry date, and issues the certificate with its issue date and its expiry where both sides can see it. Use it when running `ref/prt/certify-a-partner` and `ref/rev/certify-a-seller`. license: CC-BY-4.0 metadata: agent: certification-check agent-version: "1" --- # Record and issue what was passed ## What it does Records the result with the standard version, the assessors, the evidence and the expiry date, and issues the certificate with its issue date and its expiry where both sides can see 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. - **Issue the Certificates** - `ref/prt/certify-a-partner`, activity 14 - [Certify a Partner](../../../../processes/partners/certify-a-partner.md) - **Record the Certification** - `ref/rev/certify-a-seller`, activity 16 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) ## What to record Per person: the request, naming who is put forward, for which job and by when. The parts that must be proved, each with the reason a customer would notice it. The assessment as checked, item by item, each item naming the part of the standard it tests, and every item removed with the reason. The assessors, and the conflicts that were ruled out. What the assessors were briefed on, at a version. The sitting as scheduled, including what the person was told they may bring. A score per part, with the evidence it was given on and the assessor who gave it, kept so a disputed score can be reread rather than re-argued. A score set aside for conflict is kept marked as set aside, never deleted. The signed result, who signed it, what it covers, and the condition and expiry on any conditional pass. The certificate with its issue date, requirement version and expiry. Attendance by name against the list of who must be told. The later report setting scores against what those people did. 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/report-what-the-standard-predicted/SKILL.mdReports months later whether the people it passed went on to do the job, because a standard that passes people who cannot sell is not a gate. Use it when running `ref/rev/certify-a-seller`.
--- name: report-what-the-standard-predicted description: Reports months later whether the people it passed went on to do the job, because a standard that passes people who cannot sell is not a gate. Use it when running `ref/rev/certify-a-seller`. license: CC-BY-4.0 metadata: agent: certification-check agent-version: "1" --- # Report what the standard predicted ## What it does Reports months later whether the people it passed went on to do the job, because a standard that passes people who cannot sell is not a gate. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Report What Certification Predicted** - `ref/rev/certify-a-seller`, activity 17 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) ## What to record Per person: the request, naming who is put forward, for which job and by when. The parts that must be proved, each with the reason a customer would notice it. The assessment as checked, item by item, each item naming the part of the standard it tests, and every item removed with the reason. The assessors, and the conflicts that were ruled out. What the assessors were briefed on, at a version. The sitting as scheduled, including what the person was told they may bring. A score per part, with the evidence it was given on and the assessor who gave it, kept so a disputed score can be reread rather than re-argued. A score set aside for conflict is kept marked as set aside, never deleted. The signed result, who signed it, what it covers, and the condition and expiry on any conditional pass. The certificate with its issue date, requirement version and expiry. Attendance by name against the list of who must be told. The later report setting scores against what those people did. 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/run-the-sitting/SKILL.md"Runs the sitting itself: the written exam against the question set at its version, the product, pricing and market questions answered cold, and the practical watched while it is done, so somebody is checked before a customer is the one finding out. Use it when running `ref/prt/certify-a-partner`, `ref/rev/certify-a-seller` and `ref/rev/onboard-a-seller`."
--- name: run-the-sitting description: "Runs the sitting itself: the written exam against the question set at its version, the product, pricing and market questions answered cold, and the practical watched while it is done, so somebody is checked before a customer is the one finding out. Use it when running `ref/prt/certify-a-partner`, `ref/rev/certify-a-seller` and `ref/rev/onboard-a-seller`." license: CC-BY-4.0 metadata: agent: certification-check agent-version: "1" --- # Run the sitting ## What it does Runs the sitting itself: the written exam against the question set at its version, the product, pricing and market questions answered cold, and the practical watched while it is done, so somebody is checked before a customer is the one finding out. ## Where it happens The agent does this in four activities across three reference processes. Each one names the activity as that process words it. - **Assess the Practical** - `ref/prt/certify-a-partner`, activity 11 - [Certify a Partner](../../../../processes/partners/certify-a-partner.md) - **Prove the Parts that Matter** - `ref/rev/onboard-a-seller`, activity 14 - [Onboard a Seller](../../../../processes/revenue-operations/onboard-a-seller.md) - **Run the Knowledge Check** - `ref/rev/certify-a-seller`, activity 9 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) - **Sit the Written Exam** - `ref/prt/certify-a-partner`, activity 10 - [Certify a Partner](../../../../processes/partners/certify-a-partner.md) ## What to record Per person: the request, naming who is put forward, for which job and by when. The parts that must be proved, each with the reason a customer would notice it. The assessment as checked, item by item, each item naming the part of the standard it tests, and every item removed with the reason. The assessors, and the conflicts that were ruled out. What the assessors were briefed on, at a version. The sitting as scheduled, including what the person was told they may bring. A score per part, with the evidence it was given on and the assessor who gave it, kept so a disputed score can be reread rather than re-argued. A score set aside for conflict is kept marked as set aside, never deleted. The signed result, who signed it, what it covers, and the condition and expiry on any conditional pass. The certificate with its issue date, requirement version and expiry. Attendance by name against the list of who must be told. The later report setting scores against what those people did. 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-against-a-written-pass-line/SKILL.mdScores every part against the pass line as it was written before use, and keeps the evidence beside each score with the assessor who gave it, so a disputed score can be reread rather than re-argued. Use it when running `ref/rev/certify-a-seller`.
--- name: score-against-a-written-pass-line description: Scores every part against the pass line as it was written before use, and keeps the evidence beside each score with the assessor who gave it, so a disputed score can be reread rather than re-argued. Use it when running `ref/rev/certify-a-seller`. license: CC-BY-4.0 metadata: agent: certification-check agent-version: "1" --- # Score against a written pass line ## What it does Scores every part against the pass line as it was written before use, and keeps the evidence beside each score with the assessor who gave it, so a disputed score can be reread rather than re-argued. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Score Against the Standard** - `ref/rev/certify-a-seller`, activity 13 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) ## What to record Per person: the request, naming who is put forward, for which job and by when. The parts that must be proved, each with the reason a customer would notice it. The assessment as checked, item by item, each item naming the part of the standard it tests, and every item removed with the reason. The assessors, and the conflicts that were ruled out. What the assessors were briefed on, at a version. The sitting as scheduled, including what the person was told they may bring. A score per part, with the evidence it was given on and the assessor who gave it, kept so a disputed score can be reread rather than re-argued. A score set aside for conflict is kept marked as set aside, never deleted. The signed result, who signed it, what it covers, and the condition and expiry on any conditional pass. The certificate with its issue date, requirement version and expiry. Attendance by name against the list of who must be told. The later report setting scores against what those people did. 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-sitting-must-prove/SKILL.mdTakes in who is put forward, for which job and by when, names the parts of the job a customer would notice, and checks the assessment item by item against the standard it claims to test. Use it when running `ref/rev/certify-a-seller`.
--- name: settle-what-a-sitting-must-prove description: Takes in who is put forward, for which job and by when, names the parts of the job a customer would notice, and checks the assessment item by item against the standard it claims to test. Use it when running `ref/rev/certify-a-seller`. license: CC-BY-4.0 metadata: agent: certification-check agent-version: "1" --- # Settle what a sitting must prove ## What it does Takes in who is put forward, for which job and by when, names the parts of the job a customer would notice, and checks the assessment item by item against the standard it claims to test. An item that tests something the standard does not name is changed rather than scored. ## Where it happens The agent does this in three activities across one reference process. Each one names the activity as that process words it. - **Check the Assessment Is Fair** - `ref/rev/certify-a-seller`, activity 5 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) - **Name What Must Be Proved** - `ref/rev/certify-a-seller`, activity 2 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) - **Take in the Request to Certify** - `ref/rev/certify-a-seller`, activity 1 - [Certify a Seller](../../../../processes/revenue-operations/certify-a-seller.md) ## What to record Per person: the request, naming who is put forward, for which job and by when. The parts that must be proved, each with the reason a customer would notice it. The assessment as checked, item by item, each item naming the part of the standard it tests, and every item removed with the reason. The assessors, and the conflicts that were ruled out. What the assessors were briefed on, at a version. The sitting as scheduled, including what the person was told they may bring. A score per part, with the evidence it was given on and the assessor who gave it, kept so a disputed score can be reread rather than re-argued. A score set aside for conflict is kept marked as set aside, never deleted. The signed result, who signed it, what it covers, and the condition and expiry on any conditional pass. The certificate with its issue date, requirement version and expiry. Attendance by name against the list of who must be told. The later report setting scores against what those people did. 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 7 roles, and the access level beside each one is the most the work needs.
certification-record · writelearning-system · writeenablement-library · readcrm · writequoting-system · writepartner-portal · writeanalytics · 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": {
"certification-record": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/certification-record"
},
"learning-system": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/learning-system"
},
"enablement-library": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/enablement-library"
},
"crm": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/crm"
},
"quoting-system": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/quoting-system"
},
"partner-portal": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/partner-portal"
},
"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.
Where it works
5 processes, 19 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.
Certify a Partnerref/prt/certify-a-partner · 16 activities
Sit the Written Exam · Assess the Practical · Decide Pass or Fail · Issue the Certificates
Run a First Deal Togetherref/prt/run-a-first-deal-together · 16 activities
Decide They Can Do It Unaided
Certify a Sellerref/rev/certify-a-seller · 17 activities
Take in the Request to Certify · Name What Must Be Proved · Check the Assessment Is Fair · Name the Assessors · Brief the Assessors · Schedule the Sitting · Run the Knowledge Check · Score Against the Standard · Decide Pass or Retake · Record the Certification · Report What Certification Predicted
Onboard a Sellerref/rev/onboard-a-seller · 17 activities
Prove the Parts that Matter
Roll Out a Changeref/rev/roll-out-a-change · 17 activities
Check Who Attended · Check That It Landed
The records it leaves
What it writes down
Per person: the request, naming who is put forward, for which job and by when. The parts that must be proved, each with the reason a customer would notice it. The assessment as checked, item by item, each item naming the part of the standard it tests, and every item removed with the reason. The assessors, and the conflicts that were ruled out. What the assessors were briefed on, at a version. The sitting as scheduled, including what the person was told they may bring. A score per part, with the evidence it was given on and the assessor who gave it, kept so a disputed score can be reread rather than re-argued. A score set aside for conflict is kept marked as set aside, never deleted. The signed result, who signed it, what it covers, and the condition and expiry on any conditional pass. The certificate with its issue date, requirement version and expiry. Attendance by name against the list of who must be told. The later report setting scores against what those people did.
Where it stops
What it will not do
Does not teach, and does not write the courses or the questions, which is the curriculum-manager agent's work. Does not write the standard it scores against, which belongs to the standards-keeper agent, and an assessment that tests something the standard does not name is changed rather than scored. Does not decide who is put forward or run the ramp around them, which is the onboarding-manager agent's job. Does not lower a pass mark, move it after a sitting, or issue a certification on its own authority.
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: certification-check 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.
- Advisors AI Store Readiness Check MCP server · advisorsai.ai · hosted · mirrored from the MCP registry One read-only public-page readiness check with a signed, recheckable evidence receipt.
- Vigilis MCP server · piyushpathakqa.github.io · run it yourself · mirrored from the MCP registry Gate AI-written tests: heal safe drift, refuse to hide real bugs, sign a verifiable receipt.
- rigour MCP server · rigour-labs.github.io · run it yourself · mirrored from the MCP registry Quality gates for AI agents. Lint, test, build checks with memory persistence.