aibast-agents-library / vault
llms.txt8 starsChanged 3 months agoMITView on GitHub
The reference vault for the RAPP production guide in the AIBAST Agents Library: every step of the 14-step guide expanded, and the RAPP/1 protocol it rests on, as linked Markdown notes. Notes that restate the RAPP/1 specification are pinned to upstream commit 0b24e7d60786d39d97265338add9ca784bf5a7ab (MIT; see NOTICE.md).
In this repository539 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- community_rapp/project/CLAUDE.mdCLAUDE.md
- rapp_ai/CLAUDE.mdCLAUDE.md
- rapp_brainstem/CLAUDE.mdCLAUDE.md
- community_rapp/project/.github/copilot-instructions.mdCopilot instructions
- .github/copilot-instructions.mdCopilot instructions
- rapp_ai/.github/copilot-instructions.mdCopilot instructions
- docs/vault/llms-full.txtllms.txt
- csv-surgeonSkill
- json-doctorSkill
- log-detectiveSkill
- markdown-medicSkill
- pii-scoutSkill
- generate-demoSkill
- community_rappSkill
- generate-demoSkill
- rapp-brainstemSkill
- aibast-easy-mode-brainstemSkill
- aibast-easy-mode-copilotSkill
- aibast-staging-ring-testSkill
- account-intelligence-account-overviewSkill
- account-intelligence-competitive-intelSkill
- account-intelligence-executive-briefingSkill
- account-intelligence-risk-assessmentSkill
- account-intelligence-stakeholder-mapSkill
- account-intelligence-value-messagingSkill
- escalation-routingSkill
- follow-up-planSkill
- handle-inquirySkill
- interaction-summarySkill
- knowledge-searchSkill
- prepare-actionsSkill
- recommend-resolutionSkill
- satisfaction-surveySkill
- benefits-summarySkill
- health-insuranceSkill
- leave-balanceSkill
- parental-leaveSkill
- remindersSkill
- remote-workSkill
- submit-time-offSkill
- time-off-checkSkill
- asset-maintenance-forecast-asset-healthSkill
- asset-maintenance-forecast-budget-projectionSkill
- asset-maintenance-forecast-maintenance-forecastSkill
- asset-maintenance-forecast-work-order-planSkill
- compare-plansSkill
- deadline-remindersSkill
- document-checklistSkill
- election-draftSkill
- enrollment-windowSkill
- life-event-changeSkill
- provider-networkSkill
- approval-routeSkill
- content-scanSkill
- draft-briefSkill
- figure-checkSkill
- gap-reportSkill
- list-templatesSkill
- proactive-applicant-updatesSkill
- The first 60 shown.
The file
Files
# RAPP/1 Reference Vault > The reference vault for the RAPP production guide in the AIBAST Agents Library: every step of the > 14-step guide expanded, and the RAPP/1 protocol it rests on, as linked Markdown notes. Notes that > restate the RAPP/1 specification are pinned to upstream commit 0b24e7d60786d39d97265338add9ca784bf5a7ab (MIT; see NOTICE.md). - [Production guide](https://microsoft.github.io/aibast-agents-library/docs/vault/../rapp-guide.html): the 14-step RAPP production process - [All notes as one file](https://microsoft.github.io/aibast-agents-library/docs/vault/llms-full.txt) - [Obsidian vault download](https://microsoft.github.io/aibast-agents-library/docs/vault/rapp1-vault.zip) ## Vault - [Welcome](https://microsoft.github.io/aibast-agents-library/docs/vault/README.md): What this vault is, who it is for, and where to start. - [Index](https://microsoft.github.io/aibast-agents-library/docs/vault/_index.md): Every note in the vault, grouped by section. - [Notice](https://microsoft.github.io/aibast-agents-library/docs/vault/NOTICE.md): Where the vault's content and viewer come from, and the licences that apply. ## Start Here - [How to Use This Vault](https://microsoft.github.io/aibast-agents-library/docs/vault/Start%20Here/How%20to%20Use%20This%20Vault.md): How do I read, search, download and share this vault? - [Reading Paths](https://microsoft.github.io/aibast-agents-library/docs/vault/Start%20Here/Reading%20Paths.md): Which notes should I read, in what order, for my role? ## Production Guide - [Production Guide Overview](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Production%20Guide%20Overview.md): What the 14-step RAPP production process is, who runs each step, and where the six human quality gates sit. - [Step 01 Discovery Call](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2001%20Discovery%20Call.md): How to run the discovery call whose transcript becomes the foundation for every later step. - [Step 02 Transcript Analysis](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2002%20Transcript%20Analysis.md): How the transcript is turned into structured project JSON and audited at Quality Gate 1. - [Step 03 Generate MVP Poke](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2003%20Generate%20MVP%20Poke.md): What an MVP poke is, what goes in it, and how it is sent to the customer before any code is written. - [Step 04 Customer Validation](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2004%20Customer%20Validation.md): How the customer signs off on the MVP scope at Quality Gate 2, and what to do for each possible answer. - [Step 05 Generate Agent Code](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2005%20Generate%20Agent%20Code.md): How the locked MVP becomes Python agent files that follow the BasicAgent pattern. - [Step 06 Code Quality Review](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2006%20Code%20Quality%20Review.md): What a reviewer checks in generated agent code at Quality Gate 3 before it is deployed. - [Step 07 Deploy Prototype](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2007%20Deploy%20Prototype.md): How reviewed agents get running locally, in Azure, and in Copilot Studio, and how to verify the deployment. - [Step 08 Demo Review](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2008%20Demo%20Review.md): How the customer first sees the working prototype, and how Quality Gate 4 decides what happens next. - [Step 09 Generate Video Demo](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2009%20Generate%20Video%20Demo.md): How to turn the approved prototype into a produced, narrated demo for executives. - [Step 10 Final Demo Review](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2010%20Final%20Demo%20Review.md): How the customer approves the produced video demo for executive sharing at Quality Gate 5. - [Step 11 Production Planning](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2011%20Production%20Planning.md): How the prototype's requirements for security, compliance, and rollout are defined before production. - [Step 12 Security Sign-off](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2012%20Security%20Sign-off.md): How IT, Security, and Compliance formally approve production deployment at Quality Gate 6. - [Step 13 Production Deployment](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2013%20Production%20Deployment.md): How the approved system is deployed to the customer's environment and rolled out to pilot users. - [Step 14 Post-Deployment Review](https://microsoft.github.io/aibast-agents-library/docs/vault/Production%20Guide/Step%2014%20Post-Deployment%20Review.md): How the pilot is measured against the MVP targets and how Phase 2 is planned. ## Protocol - [Canonicalization](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/Canonicalization.md): How does rapp/1 turn any JSON value into one exact byte string? - [Conformance Classes](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/Conformance%20Classes.md): What does an implementation have to do to claim it conforms to rapp/1, and what extra rules apply to the Grail kernel and production operations? - [Content Addressing](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/Content%20Addressing.md): How does rapp/1 hash things, and why is every hash tagged with a space? - [Identity and the Rappid](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/Identity%20and%20the%20Rappid.md): What is a rappid, how is it minted, and when may it ever change? - [Protocol at a Glance](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/Protocol%20at%20a%20Glance.md): What is rapp/1, and what does it promise in one page? - [Protocol Layers](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/Protocol%20Layers.md): How is rapp/1 split into layers, and what may each layer do? - [Security Considerations](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/Security%20Considerations.md): Which threats does rapp/1 defend against, which ones do hashes alone not stop, and what control covers each? - [The Egg](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/The%20Egg.md): How does rapp/1 package a whole agent, application, or space so anyone can verify it before using it? - [The Frame](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/The%20Frame.md): What is a rapp/1 frame, what are its eleven fields, and how is it verified? - [The Registry](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/The%20Registry.md): What is the estate registry, why is it the root of trust, and how does a consumer know it is genuine and current? - [The Wire](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/The%20Wire.md): How do RAPP systems talk to each other, and what exactly does POST /chat accept and return? - [Trust and Signatures](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/Trust%20and%20Signatures.md): What does a rapp/1 signature prove, how is it checked, and how do key rotation and revocation work? - [Versioning and Evolution](https://microsoft.github.io/aibast-agents-library/docs/vault/Protocol/Versioning%20and%20Evolution.md): How does rapp/1 change over time without breaking verification of anything already published? ## Operations - [RAPP CICD](https://microsoft.github.io/aibast-agents-library/docs/vault/Operations/RAPP%20CICD.md): How does one exact AI release candidate earn promotion toward production under rapp/1? - [RAPP Deploy](https://microsoft.github.io/aibast-agents-library/docs/vault/Operations/RAPP%20Deploy.md): How does a qualified rapp/1 release reach users safely and stay provably healthy afterward? - [RAPP Work](https://microsoft.github.io/aibast-agents-library/docs/vault/Operations/RAPP%20Work.md): How does rapp/1 bind a work organization to its catalog, releases, migrations, and rollback targets without granting hidden authority? ## Building - [Conformance and Drift](https://microsoft.github.io/aibast-agents-library/docs/vault/Building/Conformance%20and%20Drift.md): How do you prove an implementation conforms to rapp/1, and how do you find and fix drift in real artifacts? - [Distros and Pinning](https://microsoft.github.io/aibast-agents-library/docs/vault/Building/Distros%20and%20Pinning.md): How does this library ship the Brainstem kernel pinned by a species contract and drift-ledgered, with every difference recorded in `rapp/KERNEL-DRIFT.md`, and its own catalog around it? - [Implementing the Protocol](https://microsoft.github.io/aibast-agents-library/docs/vault/Building/Implementing%20the%20Protocol.md): In what order should you build a rapp/1 implementation, and how do you know it is done? - [The Agent Contract](https://microsoft.github.io/aibast-agents-library/docs/vault/Building/The%20Agent%20Contract.md): What exactly must a Brainstem agent file look like, and how does it reach the user through POST /chat? - [The Brainstem Kernel](https://microsoft.github.io/aibast-agents-library/docs/vault/Building/The%20Brainstem%20Kernel.md): What is the Brainstem kernel this library ships, and what does it guarantee to every agent and host? ## ms-rapp-1 - [Microsoft Foundry](https://microsoft.github.io/aibast-agents-library/docs/vault/ms-rapp-1/Microsoft%20Foundry.md): How is ms-rapp/1 used in Microsoft Foundry, and what does a Foundry agent see? - [ms-rapp-1 Overview](https://microsoft.github.io/aibast-agents-library/docs/vault/ms-rapp-1/ms-rapp-1%20Overview.md): What is ms-rapp/1, and how does it stay backward compatible with rapp/1 and the Brainstem? - [One Concept Across the Stack](https://microsoft.github.io/aibast-agents-library/docs/vault/ms-rapp-1/One%20Concept%20Across%20the%20Stack.md): How does one Step file show up in Copilot Studio, Microsoft 365 Copilot, Foundry, the Agent Framework and the Agents SDK? - [RAPP Runtime](https://microsoft.github.io/aibast-agents-library/docs/vault/ms-rapp-1/RAPP%20Runtime.md): What is RAPP Runtime, and how does each of its pieces map to the Brainstem? - [Run Receipts and Change Control](https://microsoft.github.io/aibast-agents-library/docs/vault/ms-rapp-1/Run%20Receipts%20and%20Change%20Control.md): How does ms-rapp/1 keep an AI step auditable, and how does a new Step get into production? - [Steps](https://microsoft.github.io/aibast-agents-library/docs/vault/ms-rapp-1/Steps.md): What does a Step file look like, and why does it load unchanged in a Brainstem? ## Reference - [Failure Atlas](https://microsoft.github.io/aibast-agents-library/docs/vault/Reference/Failure%20Atlas.md): When a frame is refused, which check failed and what usually caused it? - [Glossary](https://microsoft.github.io/aibast-agents-library/docs/vault/Reference/Glossary.md): What does each rapp/1 term mean, in plain words? - [Sources and Provenance](https://microsoft.github.io/aibast-agents-library/docs/vault/Reference/Sources%20and%20Provenance.md): Where does this vault's protocol content come from, and which text is authoritative?
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

