Campaign Hub

A public library of agentic campaign specimens. Each one says what it is, when to use it, when not to, and how a stranger can tell a broken copy from a good one.

Shelves

Start here

Ten specimens from a blank page to a campaign a stranger can rerun, including the limit on what this library will hold.

  1. A stranger reuses one — The finish line for a single specimen: someone who does not know the author can reuse it, without asking a question the specimen should have answered.
  2. Specimen — One term, one definition, one home. A specimen is a reusable artifact with a kind, a license, a time to use it, a time not to, and a check. It is not a rating and it is not a run.
  3. The four strata — Piece, structure, proof, and sense. Every kind belongs to exactly one stratum. The home of that map, including the four sentences, is the kinds file.
  4. Verify before you claim — A standing rule: a claim that work is done, fixed, or published is allowed only when the speaker can point at an observation a stranger can rerun.
  5. The four-field envelope — A bead with four fields: the deliverable, how it is checked, what it may change, and what it must not touch. A missing check leaves the bead unfinished.
  6. A check that can fail — An eval a stranger can run: a bead check that exits non-zero when the how_checked field is missing or empty. A file that merely exists is not a pass.
  7. Name the consumer — A loop record that names the producer, the output, and the reader of that output in one place. A directory is storage, not a reader.
  8. A person publishes — A publish hold that opens only when a person approves. A sensor may advise. A score does not open the hold, and it does not change the kind.
  9. A campaign that ships one specimen — A one-page campaign whose product is a single specimen. It has one phase, one bead, a human publish gate, and a named reader.
  10. No attack procedures — The library does not hold procedures, commands, parameters, or payloads for compromise, credential theft, malware, or availability attacks. Naming a failure and teaching a defense are in scope. A working payload is not.

Newest

How to file a specimen

SkillpieceCC-BY-4.0

A skill an agent follows when asked to file one reusable artifact on the public shelf. It names the situation, the steps, and a refusal. It is a procedure, not …

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

skill filing refusal procedure

Defect pass

PromptpieceCC-BY-4.0

A prompt that judges one piece of work against a written checklist, lists defects, and stops. It does not rewrite the work it was asked to judge.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

prompt review defects checklist

Verify before you claim

RulepieceCC-BY-4.0

A standing rule: a claim that work is done, fixed, or published is allowed only when the speaker can point at an observation a stranger can rerun.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

rule verification claims evidence

A short public seat

Token zeropieceCC-BY-4.0

Sixteen lines that are present at the start of every turn: one home for each fact, verify before a claim, and ask before a destructive act. Public teaching text…

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

token0 seat standing-text boundary

Record, do not publish

HookpieceCC-BY-4.0

A small end-of-turn program that records that the turn ended. It refuses to publish, send, delete, or decide that the work succeeded.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

hook record side-effects turn-end

A boot note is data

Context injectionpieceCC-BY-4.0

A named slot that inserts a boot note into the window between markers. The note is data inside a boundary. It does not outrank the seat.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

context injection boundary boot-note

Read-only verifier

SubagentpieceCC-BY-4.0

A bounded helper that reads one artifact, runs the check that artifact names, and reports pass or fail. It has no write tools and no publish tools.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

subagent verifier read-only stop

Intake clerk

BotpieceCC-BY-4.0

A standing clerk that accepts a submission into a pending queue or refuses it. Approval is a different role. The clerk cannot approve.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

bot queue pending intake

Submit, moderate, read

WorkflowpieceCC-BY-4.0

The library spine as an ordered path: a person submits, a person publishes, and a stranger reads and copies. A sensor may advise the middle step and cannot publ…

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

workflow submit publish read

A stranger reuses one

GoalpieceCC-BY-4.0

The finish line for a single specimen: someone who does not know the author can reuse it, without asking a question the specimen should have answered.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

goal reuse stranger finish

One specimen in the window

Context windowpieceCC-BY-4.0

A one-shot window that holds exactly one specimen and one question: what you would change to reuse it, and what you would leave.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

window one-shot reuse boundary

The four-field envelope

BeadstructureCC-BY-4.0

A bead with four fields: the deliverable, how it is checked, what it may change, and what it must not touch. A missing check leaves the bead unfinished.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

bead envelope scope check

A campaign that ships one specimen

CampaignstructureCC-BY-4.0

A one-page campaign whose product is a single specimen. It has one phase, one bead, a human publish gate, and a named reader.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

campaign specimen gate reader

Four nodes

GraphstructureCC-BY-4.0

A campaignhub.graph/v1 graph of four nodes: apex, bead, verifier, and principal. Edges say who composes, who checks, and who guards the publish gate.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

graph nodes edges gate

Name the consumer

LoopstructureCC-BY-4.0

A loop record that names the producer, the output, and the reader of that output in one place. A directory is storage, not a reader.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

loop consumer output reader

A person publishes

GatestructureCC-BY-4.0

A publish hold that opens only when a person approves. A sensor may advise. A score does not open the hold, and it does not change the kind.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

gate publish person sensor

Markdown carrier

AdapterstructureCC-BY-4.0

How to carry any specimen as one markdown file: short front matter for the fields, then the body. The carrier adds no runtime, server, or package.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

adapter markdown front-matter carrier

For a cold reader

HandoffstructureCC-BY-4.0

A handoff a cold reader can continue from: current state, decisions already made, the next action, and what not to retry. It assumes no memory of the author.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

handoff continuity reader state

A check that can fail

EvalproofCC-BY-4.0

An eval a stranger can run: a bead check that exits non-zero when the how_checked field is missing or empty. A file that merely exists is not a pass.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

eval check exit-code bead

An envelope that goes red

FixtureproofCC-BY-4.0

A known-bad bead that omits how it is checked, paired with the failure the bead check must produce: exit 1 and a line that names the missing observation.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

fixture known-bad bead check

Complete is a claim

Failure modeproofCC-BY-4.0

The failure of treating a file that says complete as the observation. The file can be wrong. The catch is to run the check again and report that exit code.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

failure complete claim check

An empty shelf teaches the form

RetrospectiveproofCC-BY-4.0

A retrospective of a registry that shipped a submission form and no specimens. People learned how to submit. They learned nothing about the craft. A library sta…

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

retrospective shelf library form

Four scores, two safety questions

ScorecardproofCC-BY-4.0

A scorecard of four advisory scores, two yes-or-no safety questions that can block a publish, and a kind suggestion that never overwrites the kind a person chos…

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

scorecard sensor safety kind

Icing and fuel-out

Kill pairproofCC-BY-4.0

The stop pair for this library: icing is executing an untrusted specimen or letting a sensor publish. Fuel-out is a half-applied schema with no previous release…

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

kill-pair icing fuel-out stop

One value, one source

Best practiceproofCC-BY-4.0

A practice that stayed in force because a check can see it: a fact has one home, a second copy names that home, and a reader can point at the home.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

source fact home copy

Specimen

GlossarysenseCC-BY-4.0

One term, one definition, one home. A specimen is a reusable artifact with a kind, a license, a time to use it, a time not to, and a check. It is not a rating a…

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

glossary specimen definition artifact

The four strata

GlossarysenseCC-BY-4.0

Piece, structure, proof, and sense. Every kind belongs to exactly one stratum. The home of that map, including the four sentences, is the kinds file.

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

glossary strata kinds map

Apex

Role cardsenseCC-BY-4.0

The apex decomposes the work, writes the checks, integrates, and tells the truth about what was verified. Workers produce volume inside a scope. The apex does n…

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

role apex integration checks

No attack procedures

PolicysenseCC-BY-4.0

The library does not hold procedures, commands, parameters, or payloads for compromise, credential theft, malware, or availability attacks. Naming a failure and…

Original teaching specimen written for the Campaign Hub library. Not a transcript of a private run.

policy safety defense scope