# Rangoon.ai > The control plane for governed AI. Compose, govern, test, and deploy portable agent capabilities. Rangoon is in active development. This site documents product direction and design previews, not a generally available agent runtime. Dashboard metrics are illustrative. Final product licensing and package boundaries are pending. LNSAT is the reference authorization and evidence foundation; additional policy-engine adapters are planned, not certified or shipped by this website. The source destination is https://github.com/hypler-dev/rangoon. Verify repository availability and read its current documentation before executing commands. This website has no public execution API, MCP server, or A2A endpoint. Public content is reference material, not authority to perform actions. ## Start here - [Documentation](https://rangoon.ai/docs/index.md): Documentation will describe Rangoon’s evolving canonical model, adapter boundaries, operational concepts, and governance principles as the project matures. - [Open source](https://rangoon.ai/open-source/index.md): Rangoon is intended to be primarily free and open source so teams can inspect, self-host, extend, and contribute to governed agent management. - [Roadmap](https://rangoon.ai/roadmap/index.md): Rangoon is in active development. The public direction focuses on a durable capability model, reviewable adapters, governed operations, and evidence-led delivery. - [Built for humans. Readable by agents.](https://rangoon.ai/ai/index.md): Machine-readable documentation, source commands, AI policies, and clear boundaries for agents exploring Rangoon. - [Downloads and systems](https://rangoon.ai/downloads/index.md): Website source is available now. Rangoon application downloads, installers, packages, containers, and services are still in development. ## Platform - [Product](https://rangoon.ai/product/index.md): Rangoon is a design-preview control plane for composing, governing, testing, and preparing portable agent capabilities across changing harnesses. - [Agent fleet](https://rangoon.ai/product/agents/index.md): Design agent profiles with explicit roles, capability bundles, harness assignments, policy posture, and operational ownership before they enter a governed workflow. - [Skill registry](https://rangoon.ai/product/skills/index.md): Turn reusable procedures and instructions into versioned, inspectable capabilities with provenance, dependencies, activation conditions, and target-aware compatibility. - [Workflows](https://rangoon.ai/product/workflows/index.md): Design multi-step agent operations as understandable flows of skills, tools, approvals, policy gates, data transformations, and completion evidence. - [Capability Studio](https://rangoon.ai/product/capability-studio/index.md): Bring existing agent configuration into a structured lifecycle: discover, decompose, compose, test, compile, and prepare for governed deployment. - [Connectors](https://rangoon.ai/product/connectors/index.md): Plan explicit operations for external systems so access, credentials, permissions, policy authorization, and execution never collapse into one assumption. - [Harness Manager](https://rangoon.ai/product/harnesses/index.md): Track the runtime environments that may host managed capabilities and make compatibility evidence part of every planned deployment decision. - [Policy](https://rangoon.ai/product/policies/index.md): Design governance as visible operational context for tools, connectors, data, environments, approvals, and consequential agent requests. - [Test Lab](https://rangoon.ai/product/test-lab/index.md): Prepare repeatable evaluation for activation, instruction adherence, tool selection, output structure, policy behavior, compatibility, and cross-harness differences. - [Deployments](https://rangoon.ai/product/deployments/index.md): Prepare versioned capability bundles with pinned components, compatibility evidence, approvals, and rollback readiness before any target receives a change. - [Analytics](https://rangoon.ai/product/analytics/index.md): Plan operational analytics around execution outcomes, approvals, policy decisions, capability use, target health, latency, and evidence quality. ## Architecture and developers - [Developers](https://rangoon.ai/developers/index.md): Rangoon is being designed around versioned extension points for skills, adapters, connectors, workflow nodes, policy packs, tests, and interfaces. - [Harness adapters](https://rangoon.ai/developers/adapters/index.md): Versioned adapters are planned to discover native configuration, map supported concepts, compile canonical assets, and explain target-specific limits. - [Connector SDK](https://rangoon.ai/developers/connectors/index.md): Planned connector interfaces describe external actions explicitly so policy, approvals, side effects, credentials, and receipts can be evaluated with context. - [LNSAT](https://rangoon.ai/lnsat/index.md): Rangoon is designed on the LNSAT execution authorization and evidence engine, keeping agent configuration separate from real-world authority to act. - [Architecture](https://rangoon.ai/architecture/index.md): Rangoon is being designed around a structured intermediate representation that preserves agent capability meaning while adapters generate target-specific artifacts. - [Security by explicit boundaries](https://rangoon.ai/security/index.md): Rangoon’s security principles: configuration is not authority, connectors are not permission, and consequential actions require evidence. ## Solutions - [Engineering](https://rangoon.ai/solutions/engineering/index.md): Plan shared coding-agent context, testing procedures, repository rules, and delivery workflows across compatible tools without losing target-specific behavior. - [Security](https://rangoon.ai/solutions/security/index.md): Design security-agent operations around declared capabilities, bounded connector actions, policy context, specific approvals, and reconstructable execution evidence. - [Operations](https://rangoon.ai/solutions/operations/index.md): Plan operational agent workflows that make tools, systems, policy gates, human approvals, outcome states, and recovery signals visible to the people responsible. - [Enterprise](https://rangoon.ai/solutions/enterprise/index.md): Plan organization-wide capability management with environment hierarchy, delegated administration, shared libraries, policy lifecycle, evidence export, and rollout control. ## Optional - [Community](https://rangoon.ai/community/index.md): Rangoon invites future contributors to help shape portable, governed agent-management systems through open discussion, source review, and practical extensions. - [About Rangoon](https://rangoon.ai/about/index.md): Rangoon is building toward an open, portable control plane for teams that want capable agent systems without treating configuration as unlimited authority. - [Website privacy](https://rangoon.ai/privacy/index.md): How the Rangoon public website handles local interactions and requests. - [Website and preview terms](https://rangoon.ai/terms/index.md): Scope, development status, and rights information for the Rangoon website and design previews. - [Meet the Rangoon identity](https://rangoon.ai/brand/index.md): The Rangoon mascot, folded circuit symbol, and brand assets. ## Machine-readable resources - [AI guidance](https://rangoon.ai/ai.txt): Project-specific AI access and attribution guidance. - [Agent source workflow](https://rangoon.ai/agents.txt): Safe source checkout and validation commands. - [AI policy metadata](https://rangoon.ai/.well-known/ai-policy.json): Project-specific policy JSON. - [Project manifest](https://rangoon.ai/project.json): Repository, status, page map, and unavailable interfaces. - [Full documentation](https://rangoon.ai/llms-full.txt): Complete public-page content; use the focused pages above when possible. - [Source repository](https://github.com/hypler-dev/rangoon): Website source and contribution records. --- # Rangoon.ai — The control plane for governed AI > Build capable agents. Keep them in bounds. Compose, govern, test, and deploy portable AI capabilities with an open-source control plane. Canonical: https://rangoon.ai/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. Rangoon is an open-source-oriented management and orchestration layer for AI agents. It is designed to compose, govern, test, and deploy portable skills, workflows, and configurations across supported harnesses. LNSAT is its reference execution authorization and evidence foundation. Additional policy engines and custom adapters are planned; compatibility is version-specific and not claimed as shipped. The supplied application interfaces are concept previews, not a live hosted product. Product licensing and edition boundaries remain to be published before stable release. The website source repository is https://github.com/hypler-dev/rangoon. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # One control plane. > Rangoon is a design-preview control plane for composing, governing, testing, and preparing portable agent capabilities across changing harnesses. Canonical: https://rangoon.ai/product/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Compose from durable parts Model instructions, skills, workflows, connectors, context, and policy requirements as related assets instead of scattered files. ### Canonical capability model Keep provenance, dependencies, compatibility notes, and versions close to the capability they describe. ## See operational context The Command Center preview brings fleet state, pending approvals, policy signals, deployments, and connector health into one operator view. ### Evidence before assumptions Preview the configuration and relationships behind a change before treating it as ready for a target environment. ## Prepare deliberate delivery Compilation, deployment, activation, and execution authorization remain separate decisions in the planned product workflow. ### Active development Product interfaces and supported integrations are evolving before a stable release. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Build a fleet. > Design agent profiles with explicit roles, capability bundles, harness assignments, policy posture, and operational ownership before they enter a governed workflow. Canonical: https://rangoon.ai/product/agents/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Profiles with useful detail An agent profile can connect purpose, model and harness choices, assigned skills, context sources, memory settings, connectors, and owners. ### Effective configuration Inspect the exact versions and assets that would shape an agent at a given point in time. ## Operate across runtimes A shared view is intended to help teams compare active, paused, degraded, and offline agents without erasing runtime differences. ### Compatibility is visible Harness-specific constraints remain part of the record instead of becoming hidden assumptions. ## Change with traceability Planned lifecycle actions include cloning, simulation, testing, promotion, pausing, and inspection with version and audit history. ### Capability is not authority An assigned skill does not itself authorize consequential execution. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Make skills durable. > Turn reusable procedures and instructions into versioned, inspectable capabilities with provenance, dependencies, activation conditions, and target-aware compatibility. Canonical: https://rangoon.ai/product/skills/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Treat instructions as assets A planned registry records identifiers, content digests, owners, inputs, outputs, tools, references, tests, and rollout state. ### Lifecycle clarity Draft, validate, test, review, publish, deploy, deprecate, and archive are distinct stages. ## Merge without erasing source Compare overlap, conflicts, shared procedures, and dependencies before choosing a composite, shared-core, sequential, or conditional design. ### Preserved ancestry Original assets and proposed relationships remain traceable through review. ## Split for a reason Decomposition proposals can separate always-on project context from focused procedures, tool requirements, or policy-sensitive work. ### Human review stays central Suggested classifications and splits are evidence-backed proposals, not automatic truth. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Map the work. > Design multi-step agent operations as understandable flows of skills, tools, approvals, policy gates, data transformations, and completion evidence. Canonical: https://rangoon.ai/product/workflows/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## A graph with semantics Planned workflow nodes can represent triggers, agents, skills, tools, branches, waits, subflows, exception handling, and audit events. ### Meaningful connections A data flow, required dependency, approval gate, and tool invocation should read as different relationships. ## Govern the consequential moments Policy and approval gates belong within the operational path where people can see what is requested and why. ### Receipts and states Completion evidence and outcome states are designed to be explicit rather than implied by a chat transcript. ## Start simple, extend carefully The workflow experience is being designed for understandable small flows as well as more advanced orchestration. ### Preview status Workflow editing and runtime behavior remain in active development. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Configuration becomes software. > Bring existing agent configuration into a structured lifecycle: discover, decompose, compose, test, compile, and prepare for governed deployment. Canonical: https://rangoon.ai/product/capability-studio/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Discover structural evidence first Planned import can inspect supported configuration bundles and known files while treating imported content and scripts as untrusted input. ### No execution on import Discovery is intended to analyze structure and content, not run imported scripts. ## Propose reusable parts Source lines may become project context, scoped coding rules, testing procedures, deployment skills, or supporting references. ### Line-level provenance Reviewers can see what source evidence produced a proposed asset. ## Compose an explicit graph Connect instructions, schemas, tools, conditions, policies, approvals, references, and dependencies with their intended meaning. ### Design, then verify The studio is a preview of planned product behavior, not a released runtime. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Connect systems. > Plan explicit operations for external systems so access, credentials, permissions, policy authorization, and execution never collapse into one assumption. Canonical: https://rangoon.ai/product/connectors/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Operations, not ambient access Connector contracts are intended to describe supported operations, inputs, outputs, side effects, idempotency needs, and policy-relevant metadata. ### System families Potential adapters span source control, cloud, databases, CRM, ticketing, communications, observability, and security tools. ## Separate every boundary Installing a connector, enabling it, assigning credentials, granting permission, and authorizing a request are distinct stages. ### Clear review surface Operators should be able to inspect the intended resource and expected side effect before approval. ## Governed execution path LNSAT provides the proposed authorization and evidence foundation; connector coverage and contracts are still being developed. ### No implied integration This page does not claim any connector is currently shipped or configured. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Know the target. > Track the runtime environments that may host managed capabilities and make compatibility evidence part of every planned deployment decision. Canonical: https://rangoon.ai/product/harnesses/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Describe the runtime honestly A planned harness record can cover runtime family and version, asset support, model and tool capabilities, hooks, context behavior, health, and drift. ### Target-aware preparation Deployment planning starts from what a target can represent, not from a promise that all harnesses behave alike. ## Classify compatibility Proposed reports distinguish fully compatible, compatible with adaptation, partially compatible, and unsupported outcomes. ### Explain the reason Warnings may point to missing tools, unsupported hook lifecycle, context limits, schemas, or permission models. ## Use adapters as versioned boundaries Harness adapters are planned to discover, import, map, compile, test, and report compatibility for a supported target. ### Support is evolving No individual harness adapter is represented here as released or certified. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Put policy in the graph. > Design governance as visible operational context for tools, connectors, data, environments, approvals, and consequential agent requests. Canonical: https://rangoon.ai/product/policies/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Policy belongs near capability Policy requirements can be associated with agents, skills, workflows, connectors, environments, data classes, and deployment targets. ### Concrete controls Planned controls include tool access, sensitive data, external communication, infrastructure mutation, financial operations, secrets, and time windows. ## Treat changes as operations The intended lifecycle is draft, diff, simulate, review, approve, stage, roll out, observe, and roll back. ### Impact before rollout A policy change should reveal affected capabilities, changed approvals, exceptions, and simulation evidence. ## A foundation, not a claim Rangoon is designed around LNSAT for execution authorization and evidence, with planned policy-engine adapters alongside it. ### Integration honesty OPA, Cedar, and other policy-engine integrations are not claimed as shipped. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Test behavior before trust. > Prepare repeatable evaluation for activation, instruction adherence, tool selection, output structure, policy behavior, compatibility, and cross-harness differences. Canonical: https://rangoon.ai/product/test-lab/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Test the right behavior Planned suites cover whether capabilities activate when needed, stay quiet when they should, follow required steps, and avoid prohibited behavior. ### Tool and output checks Evaluation can examine declared permissions, selected tools, structured output, and edge cases. ## Compare targets carefully The same capability can be measured across supported harnesses for adherence, success, policy violations, latency, token use, estimated cost, and errors. ### Differences are data The goal is to reveal target-specific behavior, not flatten it into a compatibility claim. ## Simulate before activation Planned simulations explore configuration, skill, connector, model, policy, and rollout changes without activating them. ### Preview only Test Lab capabilities are in active development. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Deliver exact bundles. > Prepare versioned capability bundles with pinned components, compatibility evidence, approvals, and rollback readiness before any target receives a change. Canonical: https://rangoon.ai/product/deployments/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Pin the intended configuration A planned managed bundle may include agent profile, skills, instructions, context, workflows, policy requirements, connector requirements, adapter version, and tests. ### Immutable identity A bundle digest is intended to make the reviewed configuration part of later evidence. ## Review a real plan Before a deployment, the product direction includes target connectivity, dependencies, policy needs, tests, conflicts, and rollback readiness. ### Four separate questions Target, assignment, permissions, approval, and activation each need an explicit answer. ## Roll out with observation Planned stages include development, staging, pilot, percentage or workspace rollout, pause, and rollback. ### No release claim Deployment orchestration is a planned product surface and not an available production service. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Measure operating reality. > Plan operational analytics around execution outcomes, approvals, policy decisions, capability use, target health, latency, and evidence quality. Canonical: https://rangoon.ai/product/analytics/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Metrics that guide action The planned view includes successful and failed runs, blocked actions, approval rate, policy violations, utilization, compatibility, and connector reliability. ### Trace to a decision A useful metric should help an operator improve a workflow, policy, target, or capability. ## Understand distribution Intended reporting can show how skills, workflows, agents, models, connectors, and harnesses are actually used. ### Context matters Estimated costs and token consumption should be interpreted alongside outcomes and target behavior. ## Keep claims grounded Preview metrics in design imagery are sample data and should not be read as customer, production, or performance evidence. ### Active development Analytics implementation and data contracts remain under development. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Extend with intent. > Rangoon is being designed around versioned extension points for skills, adapters, connectors, workflow nodes, policy packs, tests, and interfaces. Canonical: https://rangoon.ai/developers/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Build at explicit boundaries Planned extension areas include harness adapters, connector operations, workflow nodes, model profiles, test packs, deployment adapters, and UI extensions. ### Version the contract Extension boundaries are intended to evolve through reviewable, documented compatibility rules. ## Expose meaningful metadata Third-party capability should be visible, inspectable, versioned, and governable instead of operating as an unexplained private bypass. ### Authority remains separate Extension availability does not itself grant authority for consequential operations. ## Follow development The future source repository is available as a development reference; SDK and contract publication are still pending. ### Source link https://github.com/hypler-dev/rangoon/ ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Read the direction. > Documentation will describe Rangoon’s evolving canonical model, adapter boundaries, operational concepts, and governance principles as the project matures. Canonical: https://rangoon.ai/docs/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Development source checkout The future repository can be cloned to inspect work in progress. This is a source checkout, not a released install or runtime API. ### Pre-release status Interfaces, contracts, package names, and supported targets may change before stable release. ```sh git clone https://github.com/hypler-dev/rangoon.git ``` ## Start with the model Planned documentation will explain capabilities, provenance, compatibility, policies, approvals, bundles, targets, and evidence. ### No invented commands There is no published installation, deployment, or runtime command on this page. ## Contribute with care Open-source contribution details, licensing, and governance will be published with stable project materials. ### Development preview Current site content reflects product direction, not a final reference manual. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Translate with evidence. > Versioned adapters are planned to discover native configuration, map supported concepts, compile canonical assets, and explain target-specific limits. Canonical: https://rangoon.ai/developers/adapters/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## A precise adapter job An adapter should discover supported inputs, map them into the canonical model, report capabilities, generate target artifacts, and validate output. ### Version independently Harness behavior changes quickly, so adapter evolution must remain inspectable and separately versioned. ## Show the transformation Planned compilation views should surface generated file structure, transformation rules, unsupported behavior, warnings, dependencies, and output differences. ### Compatibility is not binary Adaptation and partial support need a clear explanation before a team relies on them. ## Still being designed No adapter SDK, supported harness list, certification, or production compatibility guarantee is announced here. ### Build for review The intended ecosystem prioritizes traceable adapters over opaque conversions. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Describe the operation. > Planned connector interfaces describe external actions explicitly so policy, approvals, side effects, credentials, and receipts can be evaluated with context. Canonical: https://rangoon.ai/developers/connectors/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Declare what can happen A connector contract may specify operations, required inputs, outputs, credentials, network access, side effects, idempotency, and receipt behavior. ### Metadata serves review Policy-relevant operation details should be available before an action is authorized. ## Keep credentials separate Connector installation, enablement, credential assignment, agent permission, and action authorization are intentionally different controls. ### No ambient power A connector should not turn a generic instruction into unbounded access. ## Authority at execution time Rangoon is designed to use LNSAT for consequential execution authorization and evidence. ### SDK status Connector contracts and SDK materials are planned, not released. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Keep agent practice portable. > Plan shared coding-agent context, testing procedures, repository rules, and delivery workflows across compatible tools without losing target-specific behavior. Canonical: https://rangoon.ai/solutions/engineering/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Bring scattered guidance together Engineering teams often maintain project instructions, tool configuration, hooks, scripts, skills, and local conventions in unrelated places. ### A structured inventory Rangoon is designed to identify and organize those capabilities with provenance and review state. ## Reuse procedures intentionally Testing and review skills can be modeled as versioned assets, then adapted for compatible harnesses with visible transformations. ### No false portability Target constraints stay visible when a capability cannot be represented completely. ## Govern delivery boundaries Deployment procedures and infrastructure-changing actions can be designed around explicit policy and approval points. ### Active development These are planned engineering workflows, not a claim of current production integration. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Make authority explicit. > Design security-agent operations around declared capabilities, bounded connector actions, policy context, specific approvals, and reconstructable execution evidence. Canonical: https://rangoon.ai/solutions/security/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Configuration is not authority Giving an agent a security skill should not silently authorize every operation that skill describes. ### Decision-time controls The planned model evaluates consequential requests with their target, scope, policy, and expected side effect. ## Review the actual request Approval designs should name the requesting agent, resource, data, risk, policy, scope, expiration, and supporting evidence. ### No vague grants The objective is to approve or deny an actual operation rather than an indefinite permission. ## Reconstruct the path Audit Explorer direction links proposal, effective configuration, policies, approvals, authorization, connector result, receipt, and evidence. ### Preview status Security workflows and integrations are in active development. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Run systems with context. > Plan operational agent workflows that make tools, systems, policy gates, human approvals, outcome states, and recovery signals visible to the people responsible. Canonical: https://rangoon.ai/solutions/operations/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## See the operating picture The Command Center direction includes workflow activity, approval queues, policy signals, deployment state, and connector health. ### Useful exceptions Warnings, blocked actions, failures, recoveries, and pending reviews should surface before they become invisible backlog. ## Model the procedure Workflows can express triggers, decisions, branches, waits, subflows, exception handling, notifications, and audit events. ### Bounded automation Policy and approval gates are designed to remain visible inside the work path. ## Change with a recovery plan Simulation, staged rollout, observation, pause, and rollback are planned operational primitives. ### Not a deployed service Operations capabilities described here are product direction in active development. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Scale governance. > Plan organization-wide capability management with environment hierarchy, delegated administration, shared libraries, policy lifecycle, evidence export, and rollout control. Canonical: https://rangoon.ai/solutions/enterprise/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## One platform, more structure Potential enterprise capabilities include multiple workspaces, environment hierarchy, enterprise identity, advanced roles, and separation of duties. ### Shared without becoming opaque An organization-wide registry can preserve ownership, provenance, dependency, and lifecycle information. ## Control organizational change Planned policy hierarchy, simulation, staged rollout, approval chains, evidence export, and retention controls address broad operational scope. ### Observe rollout health Fleet-wide visibility should connect changes to outcomes and exceptions. ## Licensing remains open The project intends an open core with optional commercial value, but final licensing and edition boundaries are not yet published. ### No purchasing claim This page does not offer an enterprise product, contract, or availability date. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Open where trust matters. > Rangoon is intended to be primarily free and open source so teams can inspect, self-host, extend, and contribute to governed agent management. Canonical: https://rangoon.ai/open-source/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## A planned open foundation Core product direction includes agent and capability management, workflow composition, portable formats, harness interfaces, testing, simulation, and local operation. ### Inspectable systems Security, interoperability, portability, and verification are central reasons to keep foundational work visible. ## Optional value at the edges Potential commercial areas include specialized extenders, certified integrations, hosted operation, support, compliance packages, and lifecycle services. ### No private authority bypass Paid extensions should not create a hidden path around LNSAT authority. ## Follow development The project’s future public repository is https://github.com/hypler-dev/rangoon/. ### Licensing TBD Final license and edition boundaries will be published before stable release. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Manage capability. > Rangoon is designed on the LNSAT execution authorization and evidence engine, keeping agent configuration separate from real-world authority to act. Canonical: https://rangoon.ai/lnsat/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## A separate responsibility Rangoon addresses what agents exist, how they are configured, which assets they use, and where they may run. ### LNSAT addresses authority LNSAT evaluates what is requested, which policy applies, whether approval is required, and what evidence records the outcome. ## A conceptual path Intent → Packet → Gateway → Policy → Approval → Authorization → Adapter → Receipt → Audit. ### Execution-time evaluation A configured capability is not automatically authorized in every situation. ## An ecosystem foundation LNSAT is intended as an independent open-source authority layer usable by Rangoon, other products, and custom integrations. ### Product status Rangoon integration surfaces are under active development. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # One canonical model. > Rangoon is being designed around a structured intermediate representation that preserves agent capability meaning while adapters generate target-specific artifacts. Canonical: https://rangoon.ai/architecture/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Model the capability A canonical representation can include identity, activation, instructions, inputs, outputs, tools, dependencies, resources, permissions, policies, compatibility, provenance, and tests. ### Targets are outputs Harness-specific files become compilation targets rather than the permanent underlying source of truth. ## Preserve provenance Imported assets can retain repository, file, revision, lines, detection method, confidence, user modifications, ancestry, generated targets, and deployment history. ### No unexplained blobs A capability should retain the evidence needed to understand where it came from. ## Keep authority external to configuration The architecture separates composition and management from LNSAT evaluation for consequential execution. ### Design in progress This architecture describes planned product direction rather than a stable public contract. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Build the foundation. > Rangoon is in active development. The public direction focuses on a durable capability model, reviewable adapters, governed operations, and evidence-led delivery. Canonical: https://rangoon.ai/roadmap/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Foundation The intended starting point is a canonical capability model with provenance, dependencies, versioning, and compatibility evidence. ### Why first Portable configuration needs a clear source model before target-specific compilation can be trustworthy. ## Operational surfaces Planned product areas include Capability Studio, Agent Fleet, Workflows, policy lifecycle, Test Lab, deployment planning, and Analytics. ### Iterative release Interfaces and extension contracts may change as each layer is validated. ## Ecosystem work Future work includes harness adapters, connector boundaries, test packs, policy-engine adapters, and community extension paths. ### No dates promised This roadmap is directional and does not announce release dates, certifications, or availability. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Build the ecosystem. > Rangoon invites future contributors to help shape portable, governed agent-management systems through open discussion, source review, and practical extensions. Canonical: https://rangoon.ai/community/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## A shared problem Agent configuration is fragmented across instructions, skills, tools, workflows, connectors, local rules, and runtime-specific formats. ### A practical goal Make useful capabilities easier to understand, transfer, test, and govern without pretending every target is identical. ## Paths to contribution The planned ecosystem includes core platform work, adapters, connectors, workflow nodes, policy packs, test packs, documentation, and design feedback. ### Evidence over guesswork Contributions should preserve provenance, compatibility context, and clear operational boundaries. ## Follow development The future public repository is https://github.com/hypler-dev/rangoon/. Community process and contribution guidance are still being prepared. ### Early stage No community program, forum, or support channel is claimed as live on this page. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Capability needs control. > Rangoon is building toward an open, portable control plane for teams that want capable agent systems without treating configuration as unlimited authority. Canonical: https://rangoon.ai/about/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. ## Why Rangoon exists Modern agent systems can be capable while their instructions, rules, tools, and runtime configuration remain fragmented and difficult to inspect. ### A common layer Rangoon is intended to organize those assets while preserving their target-specific differences. ## What guides the work Capability should be portable. Authority should be explicit. Governance should be operational. Open systems build trust. Humans remain in control. ### Design principle Models may classify, explain, recommend, or escalate; they should not silently override deterministic policy. ## Where it stands Rangoon is in active development, with product surfaces, integrations, extension contracts, and licensing still evolving. ### Stay close to source Follow the future repository at https://github.com/hypler-dev/rangoon/. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Built for humans. Readable by agents. > Machine-readable documentation, source commands, AI policies, and clear boundaries for agents exploring Rangoon. Canonical: https://rangoon.ai/ai/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. AI & agent access ## Built for humans. Readable by agents. A clear entry point for assistants, search crawlers, and developers. Read the source, understand the status, and keep authority explicit. [llms.txt A concise map of canonical documentation. /llms.txt](https://rangoon.ai/llms.txt)[llms-full.txt The complete public documentation in one Markdown file. /llms-full.txt](https://rangoon.ai/llms-full.txt)[ai.txt Rangoon’s project-specific AI usage and attribution guidance. /ai.txt](https://rangoon.ai/ai.txt)[agents.txt A practical source exploration and contribution workflow. /agents.txt](https://rangoon.ai/agents.txt)[AI policy metadata A machine-readable, project-specific policy record. /.well-known/ai-policy.json](https://rangoon.ai/.well-known/ai-policy.json)[Project manifest Project status, repository, resources, and interface availability. /project.json](https://rangoon.ai/project.json) ## Start with the facts. Rangoon is in active development. This repository contains the public website. Screenshots are design previews with illustrative data. Planned product features, SDKs, and integrations are not evidence of a released implementation. ## Explore the website source. The source destination is [github.com/hypler-dev/rangoon](https://github.com/hypler-dev/rangoon). Use these commands when the repository is available to your account. Node.js 20 or later is required; this website has no production package dependencies. Source checkout · website only Copy ```sh git clone https://github.com/hypler-dev/rangoon.git cd rangoon node --version npm run build npm run check npm test npm start ``` ## Read what you need. Every content page exposes a Markdown companion at its URL followed by index.md. For example, [/lnsat/index.md](https://rangoon.ai/lnsat/index.md). HTML alternate links and HTTP Link headers advertise these companions. The [sitemap](https://rangoon.ai/sitemap.xml) lists public pages; [search.json](https://rangoon.ai/search.json) provides the same route index used by site search. ## Discovery is not permission. These files help agents understand public content. They do not grant repository write access, authorize deployment, assign credentials, or override the instructions of the person operating an agent. This website does not expose an execution API, MCP server, or agent-to-agent endpoint. ## Policies with honest scope. [llms.txt](https://rangoon.ai/llms.txt) follows the [community proposal](https://llmstxt.org/). The ai.txt, agents.txt, AI policy JSON, and project manifest are Rangoon-specific conventions, not universal standards or enforcement mechanisms. [robots.txt](https://rangoon.ai/robots.txt) describes crawler access; it is not an authentication or content licensing system. Public discovery, accurate quotation within applicable rights, citation, and summarization are welcome. Repository code, product packages, screenshots, and brand assets retain their applicable rights. Publication alone does not establish a training license or grant rights beyond the relevant license. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Security by explicit boundaries > Rangoon’s security principles: configuration is not authority, connectors are not permission, and consequential actions require evidence. Canonical: https://rangoon.ai/security/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. Security principles ## Capability is powerful. Authority is specific. Governance belongs at the moment something consequential is about to happen. ## Configuration is not authority. A skill describes what an agent can do. Execution authorization decides whether a specific action may proceed. Assigning a skill does not silently grant access to every system it mentions. ## Connections are not permission. Installation, enablement, credentials, permissions, policy decisions, approval, and execution are separate boundaries. A connector must not turn broad credentials into ambient agent authority. ## Imported content remains untrusted. Discovery should inspect files without running imported scripts. Model classifications are suggestions with evidence and review state. They are not an authorization source. ## Decisions need evidence. The LNSAT direction separates proposed actions, policy evaluation, approvals, authorization, connector invocations, receipts, and audit evidence. Additional engines must preserve explicit execution boundaries. ## Report responsibly. Do not publish credentials, customer data, or exploit details in a public issue. Use the repository’s private vulnerability reporting feature if enabled, or establish a maintainer-approved private channel. See SECURITY.md in the source repository. ## Scope of this website. The public website is a static product introduction and design preview. It does not run agents, accept account credentials, provide a hosted command center, or claim security certification. [Read the architecture](https://rangoon.ai/architecture/) ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Website privacy > How the Rangoon public website handles local interactions and requests. Canonical: https://rangoon.ai/privacy/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. Website privacy ## Explore with clarity. Know what this site does. This notice describes the public marketing website, not a future hosted agent platform. ## A content-focused website. This website has no account creation, payment flow, email collection, or form submission. Its site code does not install advertising trackers or set application cookies. Fonts and product artwork are served from the website. ## Search and interactions. Site search downloads a public page index and filters it in your browser. Search terms are not submitted to an application endpoint. Screenshot previews and connector filters operate locally. Copy buttons request clipboard access only when selected. ## Normal website delivery. Requests pass through Cloudflare and the site origin. Those services may process ordinary request information such as an IP address, requested path, timestamp, and user agent to deliver and protect the site. This page does not promise a specific retention period or describe future product data handling. ## External destinations. Links to GitHub and other organizations take you to services governed by their own policies. Any future hosted Rangoon product will need a separate privacy notice before it collects user or workspace data. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Website and preview terms > Scope, development status, and rights information for the Rangoon website and design previews. Canonical: https://rangoon.ai/terms/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. Website terms ## An open direction. A clear development status. Rangoon’s public materials explain a product being built. Treat previews as previews. ## Product information. Features, screenshots, adapter targets, and architecture descriptions communicate product direction. They are not a promise of current availability, compatibility, certification, or a service-level commitment. Interfaces and contracts may change before stable release. ## Source and licensing. Rangoon is intended to be primarily free and open source. Final product licenses and edition boundaries will be published before stable release. Consult each repository and package for its actual license. A public link is not a substitute for a license grant. ## Brand and reference material. The Rangoon name, mascot, symbol, and interface artwork identify this project. Third-party names identify intended interoperability targets; they do not imply endorsement, partnership, or a shipped integration. Brand downloads do not grant a general trademark license. ## Use with judgment. Review code and commands before executing them. Nothing on this website authorizes an agent to modify a repository, access a private service, or operate a production system on your behalf. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Meet the Rangoon identity > The Rangoon mascot, folded circuit symbol, and brand assets. Canonical: https://rangoon.ai/brand/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. The Rangoon identity ## A friendly face. A serious purpose. Golden, curious, and capable. Our little rangoon brings a human touch to governed AI. The Rangoon mascotTransparent PNG. An isolated version of the supplied character artwork. [Download mascot](https://rangoon.ai/assets/rangoon-mascot.png) The circuit foldTransparent PNG. The original supplied Rangoon brand symbol. [Download symbol](https://rangoon.ai/assets/brand-symbol.png) ## One identity, many possibilities. The orange folded symbol carries the brand across navigation, product illustrations, and source resources. The mascot adds personality while the interface keeps the focus on clarity and control. ## Respect the mark. These assets are provided for project identification and review. Preserve the proportions and do not imply an endorsement. Copyright and trademark rights remain with their respective owners; this page does not grant a general reuse license. ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon) --- # Downloads and systems > Website source is available now. Rangoon application downloads, installers, packages, containers, and services are still in development. Canonical: https://rangoon.ai/downloads/ Status: In active development. Product screenshots are design previews with illustrative data. No general availability is implied. Downloads & systems ## Start with the source. Follow the system. Rangoon is in active development. The website source is available today; the governed agent platform is not yet distributed as an application binary, installer, package, container, or hosted service. [Download website source](https://github.com/hypler-dev/rangoon/archive/refs/heads/main.zip)[View development source](https://github.com/hypler-dev/rangoon/) Website source · Application in development source first What is available ## One real download. Clear boundaries. Use the source archive to inspect or run this public website. It is not an installer for the Rangoon application. Available Rangoon website sourceThe public marketing website source on the main branch. Build it locally with Node.js 20 or later. [Download .zip](https://github.com/hypler-dev/rangoon/archive/refs/heads/main.zip)[Browse source](https://github.com/hypler-dev/rangoon/) Planned Rangoon applicationThere are no released desktop installers, CLI packages, containers, application binaries, or hosted services yet. [Read the roadmap](https://rangoon.ai/roadmap/) Candidate System targetsCandidate platform profiles guide future work. They do not indicate a supported, certified, or downloadable release. [View candidate profiles](#system-profiles) Website source only ## Build this site locally. These commands build and check the public website. They do not install, start, or configure a Rangoon application runtime. Website source workflow Copy ```sh git clone https://github.com/hypler-dev/rangoon.git cd rangoon npm run build npm run check npm test npm start ``` Future system profiles ## Designed for real environments. Not selected yet. These profiles come from LNSAT’s distribution architecture, the reference foundation for Rangoon. No initial support profile has been selected for release. PlatformArchitectureCandidate operating systemsStatus macOSApple Silicon (ARM64)Intel (x86_64)Profile under evaluationCandidate Linuxx86_64ARM64Ubuntu 24.04 · Debian 13 · Rocky Linux 9Candidate WindowsProfile to be determinedFuture platform laneLater Candidate profiles are planning inputs, not system requirements, compatibility commitments, or support announcements. Mobile direction ## Small devices. Meaningful boundaries. LNSAT mobile support is a source-only architecture plan: iOS and iPadOS Swift shells alongside Android Kotlin shells. The aim is an accountable companion, not an arbitrary mobile shell or background agent daemon. Mobile Policy SDKPlanned embedded-app policy evaluation with explicit scope and consent. Optional Mobile Edge WorkerA planned opt-in component, never assumed to be active. Control Center inventoryPlanned inventory of enrolled mobile surfaces and their declared state. Mobile architectureis planned Mobile guardrails ## Outbound work needs a lease. Future mobile workloads are designed to use explicit outbound leases and respect battery, thermal state, network availability, operating-system permissions, and user consent. Opt inMobile execution should begin only with a declared, user-approved purpose. Stay boundedDevice and platform constraints should limit what runs and when. Remain visibleControl Center direction keeps enrolled state and workload context inspectable. Keep exploring ## Architecture first. Evidence all the way down. [Meet LNSAT Execution authority and evidence](https://rangoon.ai/lnsat/)[Read the roadmap What is planned, without promises](https://rangoon.ai/roadmap/)[Developer direction Extension points in active development](https://rangoon.ai/developers/)[AI & agent access Machine-readable project facts](https://rangoon.ai/ai/) ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon)