Partner scout
partner-scout v1Finds candidate partners against a gap that has already been written down, and against the test that says when the gap is closed. Reads the partners, customers and suppliers already on the books before it reads the market, so a partner we already have is not missed in favour of a new name. Reads what each candidate sells, to whom and where, and scores every survivor against the closing test with the evidence line behind each score and the date that evidence was read. The search is answerable to the written gap, which is what keeps a list of interesting companies from becoming a list of candidates.
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.
Works alone for most of its activities. Runs collect-and-report over the known book and over the market search, and assessment when it scores candidates against the closing test. Takes the reading of each candidate from the researcher agent and the removals from the supplier-check agent, hands its scored list to the partner-manager agent for the cut, and hands the confirmed short list on to the partner-qualifier agent.
The plugin
What is in the package
5 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.
partner-scout · 10 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 write the gap or the test for closing it, and does not cut the short list or confirm it. Does not prove what a candidate can actually deliver, which is the partner-qualifier agent. Does not rule on overlap with our own sellers, which is the sales leader, and does not make first contact until that ruling exists. ## What always goes to a person - too few candidates left after screening - to the partner lead, with the removals and the reason for each, rather than widening the bounds quietly to fill a quota - a candidate that wants a different arrangement than the gap calls for - to the partner lead as a separate idea, while the run keeps scoring against the gap it was opened for - silence from the whole short list - to the partner lead with the dated replies and refusals, rather than approaching the same names again in warmer language - a candidate that asks for anything the confidentiality rule covers - to the partner lead, because nothing covered is said before an agreement covers 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 `partner-scout` leaves the
following behind, on every run.
The search bounds it was given, including the companies it may not
touch. Every candidate with the source it came from and the date it was
read. Every removal with the reason it was removed. Every score against
the closing test with the evidence line behind it and the date that
evidence was read, so a second person scoring the same candidate can get
the same answer or say where they differ. The approach as sent, the
date, and what each candidate was told. The replies, the refusals and
the silences, each dated.
An agent that leaves less than this does not implement `partner-scout`,
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 CRM | read | The partners, customers and suppliers already on the books are read before the market is. |
| market sources | read | Cap: reads per day, set by the adopting organization. |
| the partner record | write | Candidates, scores, refusals and silences are kept where the next search will read them. |
| the outreach channel | write | First contact goes out on the approach a person set, and only after the conflict check. |
| the contact record | write | Who replied, who declined, who never answered, and the reason each one gave. |
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": "partner-scout",
"version": "1.0.0",
"description": "The partner-scout abstract agent. Finds candidate partners against a gap that has already been written down, and against the test that says when the gap is closed.",
"author": {
"name": "AgentCatalog",
"url": "https://agentcatalog.com"
},
"homepage": "https://agentcatalog.com/abstract-agents/partner-scout",
"repository": "https://github.com/jeffrschneider/agentcatalog",
"license": "CC-BY-4.0",
"keywords": [
"abstract-agent",
"reference",
"partners",
"find-a-partner",
"abstract"
],
"extensions": {
"com.agentcatalog.agent": {
"agent": "partner-scout",
"agentVersion": 1,
"department": "partners",
"abstract": true,
"works": "Works alone for most of its activities. Runs collect-and-report over the known book and over the market search, and assessment when it scores candidates against the closing test. Takes the reading of each candidate from the researcher agent and the removals from the supplier-check agent, hands its scored list to the partner-manager agent for the cut, and hands the confirmed short list on to the partner-qualifier agent.",
"servers": [
{
"name": "the CRM",
"role": "crm",
"access": "read",
"needs": "The partners, customers and suppliers already on the books are read before the market is."
},
{
"name": "market sources",
"role": "market-sources",
"access": "read",
"needs": "Cap: reads per day, set by the adopting organization."
},
{
"name": "the partner record",
"role": "partner-record",
"access": "write",
"needs": "Candidates, scores, refusals and silences are kept where the next search will read them."
},
{
"name": "the outreach channel",
"role": "outreach-channel",
"access": "write",
"needs": "First contact goes out on the approach a person set, and only after the conflict check."
},
{
"name": "the contact record",
"role": "contact-record",
"access": "write",
"needs": "Who replied, who declined, who never answered, and the reason each one gave."
}
],
"records": [
"The search bounds it was given, including the companies it may not touch.",
"Every candidate with the source it came from and the date it was read.",
"Every removal with the reason it was removed.",
"Every score against the closing test with the evidence line behind it and the date that evidence was read, so a second person scoring the same candidate can get the same answer or say where they differ.",
"The approach as sent, the date, and what each candidate was told.",
"The replies, the refusals and the silences, each dated."
],
"escalates": [
"too few candidates left after screening - to the partner lead, with the removals and the reason for each, rather than widening the bounds quietly to fill a quota",
"a candidate that wants a different arrangement than the gap calls for - to the partner lead as a separate idea, while the run keeps scoring against the gap it was opened for",
"silence from the whole short list - to the partner lead with the dated replies and refusals, rather than approaching the same names again in warmer language",
"a candidate that asks for anything the confidentiality rule covers - to the partner lead, because nothing covered is said before an agreement covers it"
],
"not": [
"Does not write the gap or the test for closing it, and does not cut the short list or confirm it.",
"Does not prove what a candidate can actually deliver, which is the partner-qualifier agent.",
"Does not rule on overlap with our own sellers, which is the sales leader, and does not make first contact until that ruling exists."
],
"processes": [
{
"process": "ref/prt/find-a-partner",
"activities": [
"Set the Search Bounds",
"Look at Who We Already Know",
"Search the Market",
"Read Each Candidate",
"Screen Out the Impossible",
"Score the Candidates",
"Make First Contact",
"Record Interest and Refusal"
]
}
]
}
}
}
README.mdWhat the package is, and why it carries no mcp.json.
# partner-scout 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 five 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/find-a-partner`. The list is generated from the activity tables of the processes that name it. https://agentcatalog.com/abstract-agents/partner-scout
skills/approach-a-candidate-and-record-the-answer/SKILL.md"Makes first contact on the approach a person set, then keeps what came back: who replied, who declined, who never answered, and the reason each one gave. Use it when running `ref/prt/find-a-partner`."
--- name: approach-a-candidate-and-record-the-answer description: "Makes first contact on the approach a person set, then keeps what came back: who replied, who declined, who never answered, and the reason each one gave. Use it when running `ref/prt/find-a-partner`." license: CC-BY-4.0 metadata: agent: partner-scout agent-version: "1" --- # Approach a candidate and record the answer ## What it does Makes first contact on the approach a person set, then keeps what came back: who replied, who declined, who never answered, and the reason each one gave. ## Where it happens The agent does this in two activities across one reference process. Each one names the activity as that process words it. - **Make First Contact** - `ref/prt/find-a-partner`, activity 13 - [Find a Partner](../../../../processes/partners/find-a-partner.md) - **Record Interest and Refusal** - `ref/prt/find-a-partner`, activity 14 - [Find a Partner](../../../../processes/partners/find-a-partner.md) ## What to record The search bounds it was given, including the companies it may not touch. Every candidate with the source it came from and the date it was read. Every removal with the reason it was removed. Every score against the closing test with the evidence line behind it and the date that evidence was read, so a second person scoring the same candidate can get the same answer or say where they differ. The approach as sent, the date, and what each candidate was told. The replies, the refusals and the silences, each dated. 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-for-candidates/SKILL.mdLooks for companies that could close the gap. Use it when running `ref/prt/find-a-partner`.
--- name: look-for-candidates description: Looks for companies that could close the gap. Use it when running `ref/prt/find-a-partner`. license: CC-BY-4.0 metadata: agent: partner-scout agent-version: "1" --- # Look for candidates ## What it does Looks for companies that could close the gap. Reads the partners, customers and suppliers already on the books first and goes out to the market only after that, so a partner we already have is not missed in favour of a new name. ## Where it happens The agent does this in two activities across one reference process. Each one names the activity as that process words it. - **Look at Who We Already Know** - `ref/prt/find-a-partner`, activity 6 - [Find a Partner](../../../../processes/partners/find-a-partner.md) - **Search the Market** - `ref/prt/find-a-partner`, activity 7 - [Find a Partner](../../../../processes/partners/find-a-partner.md) ## What to record The search bounds it was given, including the companies it may not touch. Every candidate with the source it came from and the date it was read. Every removal with the reason it was removed. Every score against the closing test with the evidence line behind it and the date that evidence was read, so a second person scoring the same candidate can get the same answer or say where they differ. The approach as sent, the date, and what each candidate was told. The replies, the refusals and the silences, each dated. 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/read-and-screen-each-candidate/SKILL.mdReads what each candidate sells, to whom and where, then takes out the ones that cannot work and writes the reason against every removal. Use it when running `ref/prt/find-a-partner`.
--- name: read-and-screen-each-candidate description: Reads what each candidate sells, to whom and where, then takes out the ones that cannot work and writes the reason against every removal. Use it when running `ref/prt/find-a-partner`. license: CC-BY-4.0 metadata: agent: partner-scout agent-version: "1" --- # Read and screen each candidate ## What it does Reads what each candidate sells, to whom and where, then takes out the ones that cannot work and writes the reason against every removal. ## Where it happens The agent does this in two activities across one reference process. Each one names the activity as that process words it. - **Read Each Candidate** - `ref/prt/find-a-partner`, activity 8 - [Find a Partner](../../../../processes/partners/find-a-partner.md) - **Screen Out the Impossible** - `ref/prt/find-a-partner`, activity 9 - [Find a Partner](../../../../processes/partners/find-a-partner.md) ## What to record The search bounds it was given, including the companies it may not touch. Every candidate with the source it came from and the date it was read. Every removal with the reason it was removed. Every score against the closing test with the evidence line behind it and the date that evidence was read, so a second person scoring the same candidate can get the same answer or say where they differ. The approach as sent, the date, and what each candidate was told. The replies, the refusals and the silences, each dated. 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-the-survivors-against-the-closing-test/SKILL.mdScores every candidate that survives screening against the test the gap set, with the evidence line behind each score and the date that evidence was read. Use it when running `ref/prt/find-a-partner`.
--- name: score-the-survivors-against-the-closing-test description: Scores every candidate that survives screening against the test the gap set, with the evidence line behind each score and the date that evidence was read. Use it when running `ref/prt/find-a-partner`. license: CC-BY-4.0 metadata: agent: partner-scout agent-version: "1" --- # Score the survivors against the closing test ## What it does Scores every candidate that survives screening against the test the gap set, with the evidence line behind each score and the date that evidence was read. ## 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 the Candidates** - `ref/prt/find-a-partner`, activity 10 - [Find a Partner](../../../../processes/partners/find-a-partner.md) ## What to record The search bounds it was given, including the companies it may not touch. Every candidate with the source it came from and the date it was read. Every removal with the reason it was removed. Every score against the closing test with the evidence line behind it and the date that evidence was read, so a second person scoring the same candidate can get the same answer or say where they differ. The approach as sent, the date, and what each candidate was told. The replies, the refusals and the silences, each dated. 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-bounds-of-a-search/SKILL.mdReads the written gap and the test that says when the gap is closed, and turns them into the bounds the search runs inside, including the companies it may not touch. Use it when running `ref/prt/find-a-partner`.
--- name: set-the-bounds-of-a-search description: Reads the written gap and the test that says when the gap is closed, and turns them into the bounds the search runs inside, including the companies it may not touch. Use it when running `ref/prt/find-a-partner`. license: CC-BY-4.0 metadata: agent: partner-scout agent-version: "1" --- # Set the bounds of a search ## What it does Reads the written gap and the test that says when the gap is closed, and turns them into the bounds the search runs inside, including the companies it may not touch. ## 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 Search Bounds** - `ref/prt/find-a-partner`, activity 5 - [Find a Partner](../../../../processes/partners/find-a-partner.md) ## What to record The search bounds it was given, including the companies it may not touch. Every candidate with the source it came from and the date it was read. Every removal with the reason it was removed. Every score against the closing test with the evidence line behind it and the date that evidence was read, so a second person scoring the same candidate can get the same answer or say where they differ. The approach as sent, the date, and what each candidate was told. The replies, the refusals and the silences, each dated. 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.
crm · readmarket-sources · readpartner-record · writeoutreach-channel · writecontact-record · write
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": {
"crm": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/crm"
},
"market-sources": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/market-sources"
},
"partner-record": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/partner-record"
},
"outreach-channel": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/outreach-channel"
},
"contact-record": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/contact-record"
}
}
}
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
One process, 8 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.
Find a Partnerref/prt/find-a-partner · 16 activities
Set the Search Bounds · Look at Who We Already Know · Search the Market · Read Each Candidate · Screen Out the Impossible · Score the Candidates · Make First Contact · Record Interest and Refusal
The records it leaves
What it writes down
The search bounds it was given, including the companies it may not touch. Every candidate with the source it came from and the date it was read. Every removal with the reason it was removed. Every score against the closing test with the evidence line behind it and the date that evidence was read, so a second person scoring the same candidate can get the same answer or say where they differ. The approach as sent, the date, and what each candidate was told. The replies, the refusals and the silences, each dated.
Where it stops
What it will not do
Does not write the gap or the test for closing it, and does not cut the short list or confirm it. Does not prove what a candidate can actually deliver, which is the partner-qualifier agent. Does not rule on overlap with our own sellers, which is the sales leader, and does not make first contact until that ruling exists.
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: partner-scout 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.
- gapbase-mcp MCP server · thevibepreneur.github.io · run it yourself · mirrored from the MCP registry Search 474 validated startup gaps by industry, role, or keyword. Works offline.
- patent-gap MCP server · dropwatchhq.com · hosted · mirrored from the MCP registry Off-patent opportunity feed: expiring design patents, recall + supplier screened, for sellers.
- Scout Intel MCP — The Google for AI Agents MCP server · omniologynow-rgb.github.io · hosted · run it yourself · mirrored from the MCP registry The Google for AI agents — company intel, competitor tracking, market research via MCP. JSON output