Genesys Archivist MCP Server
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Genesys Archivist MCP ServerCapture the 'Order Status' flow and generate its documentation"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Genesys Archivist
Captures Genesys Cloud Architect flows and every resource they depend on, then generates business and technical documentation from that capture.
Two consumers, two guarantees:
Consumer | Gets | Guarantee |
Humans — engineers, PMs, customers | Markdown, PDF, and diagrams per flow | Every technical fact traces to source evidence; inference is labelled as inference |
Machines — a future, separate migration server | An immutable, schema-versioned capture bundle | Complete enough to rebuild the IVR on another platform, including prompt audio |
Archivist does not build that migration server. It guarantees the data contract that server will consume.
Status
Pre-implementation. Phase 0 has not run. Design and plans are complete; no production code exists yet.
The architecture in one paragraph
Two stages separated by a hard seam. Stage 1 (capture) is the only code that talks to Genesys: it discovers every flow of every type, fetches definitions, walks the resource reference graph to closure, downloads binary assets, and seals an immutable content-hashed capture bundle. Stage 2 (document) opens no sockets — it reads a bundle and produces Markdown, SVG diagrams, and PDF, with AI narration in the middle. Re-rendering documentation therefore costs zero Genesys API calls, and the bundle is a published contract rather than a disposable cache.
flowchart TD
A["AI client"] -->|MCP STDIO| B["MCP adapter"]
C["archivist CLI"] --> D["Application service"]
B --> D
D --> E["Genesys source provider"]
E --> F["Genesys Cloud"]
D --> G["Capture bundle (sealed, immutable)"]
G --> H["Normalize, analyze, document"]
H --> I["Markdown + diagrams + PDF"]
G --> J["Future migration server"]Getting started
npm install
npm run verify # format + lint + typecheck + test + schema validationThen read, in order:
CLAUDE.md — orientation for anyone (human or agent) about to write code here.
AGENTS.md — non-negotiable boundaries. Violating one is a release blocker.
The design spec — what is being built and why. Section 2 lists where it departs from the numbered blueprint docs below.
Plan 1: Foundation — twelve task-by-task TDD tasks that need no Genesys access.
Phase 0 spikes — the go/no-go gate that unblocks everything else.
Phase 0 is a go/no-go gate
Before the Genesys adapters are built, prove against a non-production organization that a read-only OAuth client can enumerate every required flow type across pages and divisions; that a source path can load and export published flows faithfully; that prompt audio downloads read-only; and that no mutation permission is required. Ten spikes, twelve kill criteria. See docs/spikes/.
Four source paths are in contention — Platform API, the Archy CLI, the Architect Scripting SDK, and manual YAML. Which one wins is an empirical result, not an assumption.
Repository layout
apps/cli archivist CLI
apps/mcp-server genesys-archivist MCP STDIO server
packages/domain contracts and DTOs. Pure: no I/O, no SDK types
packages/application use cases, run state machines, policy
packages/composition the one place adapters are wired to interfaces
packages/... adapters, capture, analysis, documentation, rendering, narrative
schemas/ versioned JSON Schema contracts
fixtures/ sanitized test fixtures. Never real customer configuration
docs/ blueprint, design spec, plans, ADRs, spikesDependency direction is enforced by ESLint, not by convention: domain imports nothing, application imports domain only, and apps/* stay thin.
Never commit
bundles/, derived/, documentation/, spike-evidence/, or any .wav / .mp3. Capture bundles are classified restricted — they contain endpoint URLs, DIDs, routing logic, data-table rows that may hold customer PII, and prompt audio. CI fails the build if any of these are tracked.
Terminology
The target is Genesys Cloud CX, and the IVR authoring product is Architect.
A flow has identifiers such as flowId and a version. Queues, prompts, data actions, schedules, and reusable flows also have identifiers. These are not secret API keys. A Genesys OAuth client_id and client_secret authenticate the integration and are the only secrets involved. The tool never enumerates hidden secrets, recovers OAuth client secrets, scrapes passwords, or bypasses Genesys permissions.
Non-goals for the first production release
Editing, publishing, deleting, or importing Genesys flows
Recovering or listing customer secrets
Reading live caller data, recordings, transcripts, or historical execution data
Query or Q&A tools over captured data
Remote HTTP hosting, git/PR automation, or a scheduling daemon
Claiming business intent that cannot be inferred from configuration
Blueprint documents
The original handoff. Still governing wherever the design spec does not override it.
File | Purpose |
Product goals, users, assumptions, scope | |
Components, packages, runtime decisions | |
Authentication, discovery, extraction, versions | |
MCP tools, resources, prompts, errors, jobs | |
Normalized flow graph, evidence, hashes | |
Document generation and grounding | |
Credentials, threats, authorization, data controls | |
Incremental updates, manifests, diffs, review | |
Bottlenecks, FMEA, degradation, kill criteria | |
Unit, integration, contract, security, chaos tests | |
Distribution and per-client configuration | |
Logs, metrics, audit, recovery, support | |
Ordered implementation plan | |
Definition of done and release gates | |
Questions for IST and required experiments | |
Official sources and research notes |
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
Generate cloud architecture diagrams, flowcharts, and sequence diagrams.
Generate AGENTS.md, AP2 compliance docs, checkout rules, debug playbook & MCP configs from any repo.
Build and manage Cloudgate workflow-APIs: controllers, actions, workflow graphs, and databases.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/mahmouddattiaa/genesys-architect-docs-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server