Skip to content

Feat: pivot the landing page to the project directory #135

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
mrsabath merged 5 commits into from
Nov 2, 2026
Merged

Feat: pivot the landing page to the project directory #135

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
83c46ce
Feat: pivot the landing page to the project directory
mrsabath Oct 30, 2026
d9ee322
Fix: match the GitHub CTA height and repair the landing page edit link
mrsabath Oct 30, 2026
075f4e3
Docs: drop the Past activities section from the landing page
mrsabath Oct 30, 2026
5869daa
Feat: route Get started to the projects, not the platform docs
mrsabath Oct 31, 2026
1ccf1a4
Docs: use Docusaurus Link in ProjectCards and hoist the repo URL
mrsabath Nov 2, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 8 additions & 3 deletions docusaurus.config.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@ const LATEST_VERSION = releasedVersions[0];
const config: Config = {
title: 'rossoctl',
tagline:
'A framework-neutral, scalable, secure platform for deploying, securing, and governing AI agents.',
'Open source infrastructure for running, connecting, and benchmarking AI agents.',
favicon: 'img/favicon.svg',

// Note: the `future.v4`/`experimental_faster` (Rspack) bundler is intentionally
Expand Down Expand Up @@ -199,8 +199,13 @@ const config: Config = {
path: 'ecosystem',
routeBasePath: '/',
sidebarPath: './sidebarsEcosystem.ts',
editUrl:
'https://github.com/rossoctl/ecosystem-guide/tree/main/content/',
// "Edit this page" must point at THIS repo: the ecosystem pages (the
// landing page among them) are authored here, not in the archived
// rossoctl/ecosystem-guide this instance was originally seeded from.
// Docusaurus appends the doc's site-relative path (e.g.
// 'ecosystem/welcome.mdx') to a string editUrl, so the base is the
// repo root — no 'ecosystem/' segment here, or it doubles.
editUrl: 'https://github.com/rossoctl/.github/tree/main/',
},
],
// Copies cortex/ (synced by scripts/sync-docs.sh) to build/cortex/, so
Expand Down
148 changes: 148 additions & 0 deletions ecosystem/welcome-platform-story.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,148 @@
---
draft: true
sidebar_label: Welcome (platform narrative, archived)
---

<!--
BACKUP of the pre-pivot landing page (the "Platform primitives / RossoCortex
data plane" narrative), kept so we can revert or mine it for copy. `draft:
true` keeps it out of the production build; it is not linked in the sidebar.
To restore, move this content back into welcome.mdx.
-->

import Link from '@docusaurus/Link';
import Accordion from '@site/src/components/Accordion';
import ArchitectureDiagram from '@site/src/components/ArchitectureDiagram';

export const cortexFunctions = [
{
number: '01',
status: 'Ready',
title: 'Agent identity',
description:
'Every agent carries a verifiable identity, so the platform knows who is acting before it decides what they may do.',
},
{
number: '02',
status: 'Ready',
title: 'Authorization and access',
description:
'Agents run with least privilege and act by delegation, never with more authority than the user and the agent both hold.',
},
{
number: '03',
status: 'beta in 0.7',
title: 'Intent-based access',
description:
"Access is tied to what the user actually asked for, so an agent that drifts off task can't quietly reuse its access for something else.",
},
{
number: '04',
status: 'beta in 0.7',
title: 'Tool semantic validation',
description:
"Tool calls are checked for correctness and meaning before they run, catching calls that are malformed or don't match the task.",
},
{
number: '05',
status: 'alpha in 0.7',
title: 'Context compaction',
description:
"The platform assembles and trims what goes into the model's context, keeping it relevant and lowering token cost.",
},
{
number: '06',
status: 'alpha in 0.7',
title: 'Data-flow analysis',
description:
'The platform tracks where the data behind a decision came from, so its provenance can gate what the agent does next.',
},
{
number: '07',
status: 'beta in 0.7',
title: 'Failure recovery',
description:
'The platform detects when an agent stalls or fails and returns it to a known state instead of leaving work half-finished.',
},
{
number: '08',
status: 'beta in 0.7',
title: 'User interaction',
description:
'For high-stakes steps, a person stays in the loop and approves before the agent acts.',
},
];

export const rossoctlServices = [
{
number: '01',
title: 'Skills',
description:
'Reusable, governed capabilities an agent can draw on, versioned and controlled rather than pasted in.',
},
{
number: '02',
title: 'Platform Tools',
description:
'Standalone tools the platform provides to strengthen agent and platform governance, security, resilience, reliability and efficiency, so each concern is handled once by the platform instead of rebuilt in every agent.',
},
{
number: '03',
title: 'Memory',
description:
"Durable memory that survives sessions and restarts, so an agent doesn't start from nothing each time.",
},
{
number: '04',
title: 'Knowledge base',
description:
'Shared, queryable knowledge agents use to ground their work in your own data.',
},
{
number: '05',
title: 'Sandboxes',
description:
'Isolated environments to run untrusted or self-modifying agents without risking the host or other tenants.',
},
];

<div className="rosso-landing">

# Platform primitives for trustworthy AI agents

<div className="rosso-intro">

rossoctl is a set of platform primitives for agent security, resilience, reliability and efficiency that
platform owners can build on. It comes in three parts: RossoCortex, a data plane that mediates

@aslom aslom Nov 2, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[low] This is the sentence @evaline-ju asked about on the first pass ("confused
what the 'three parts' refer to with the various commas"). The comma-spliced list —
RossoCortex, a data plane that mediates what agents do, a set of services agents can call and a set of tools... — reads as four items rather than three because the
appositive isn't set off.

It's archived, draft: true, build-excluded prose, so it genuinely does not matter
here. Raising it only because the same sentence is being added as live prose in
rossoctl/rossoctl#2662 (INSTALLER.md:3), where it is reader-facing. Fixing it
there is the useful move; semicolons already do the job in the 2662 variant, which is
most of the way there.

@mrsabath mrsabath Nov 2, 2026

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Left verbatim on purpose — welcome-platform-story.mdx is the archived draft: true backup of the pre-pivot page, so it stays byte-identical to what was live. The reader-facing variant of this sentence is fixed in rossoctl/rossoctl#2662 (1031f192): the appositive is parenthesized and the intro paragraph is split.

what agents do, a set of services agents can call and a set of tools and frameworks for observability,
security, governance and administration. It's open source, framework-neutral,
Comment on lines +116 to +118

@evaline-ju evaline-ju Oct 30, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm confused what the "three parts" refer to with the various commas, especially assuming "data plane.." is meant to describe RossoCortex rather than a separate "part"

@mrsabath mrsabath Nov 2, 2026

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good catch — that sentence now only lives in the archived draft: true backup of the old page, which stays verbatim. The live copy in rossoctl/rossoctl#2662 reads "RossoCortex (a data plane); a set of callable services; and tooling …" with the paragraph split (1031f192).

and built on open standards, supporting A2A and MCP.

</div>

<div className="margin-top--md margin-bottom--lg">
{/* Primary CTAs point at the project repos (docs hidden pre-launch). */}
<Link className="button button--lg cta-primary margin-right--sm" href="/docs/">Get started</Link>
</div>

<ArchitectureDiagram />

## The problem

AI agents don't behave like the cloud applications we know how to run. They decide which tools to call while they run, they don't cleanly separate data from instructions, their goals drift as context changes, and they can't reliably report what they did. Those traits make the ordinary demands of production hard to meet: failure recovery, admission control, security, and resource management.

Agent systems fail in ways enterprise architecture wasn't built for, so they need a platform to supply the guarantees they can't supply themselves. Kubernetes solved the same kind of problem for cloud software by separating application logic from placement, scaling, and recovery. An agent platform needs an analogous split: decouple agent logic from the guarantees around admission, isolation, and failure recovery that production demands.

## RossoCortex, the data plane

RossoCortex is an intercept. It sits transparently between an agent and everything external it touches: models, tools, users, and other agents. From that one vantage point it enforces the guarantees agents can't enforce for themselves. It is flexible by design: it works with different agent types, including black-box harnesses, integrates with different sandboxes, and works across container network choices.

<Accordion items={cortexFunctions} />

## Rossoctl services

The services are the building blocks agents need to do real work: skills, tools, memory, knowledge bases, and sandboxes. The focus is on capabilities that lower the risk of agentic workloads and use resources more efficiently.

<Accordion items={rossoctlServices} />

</div>
138 changes: 60 additions & 78 deletions ecosystem/welcome.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,138 +5,120 @@ title: Welcome to rossoctl
---

import Link from '@docusaurus/Link';
import Accordion from '@site/src/components/Accordion';
import ArchitectureDiagram from '@site/src/components/ArchitectureDiagram';
import ProjectCards from '@site/src/components/ProjectCards';

export const cortexFunctions = [
export const coreProjects = [
{
number: '01',
status: 'Ready',
title: 'Agent identity',
name: 'Moca',
repo: 'rossoctl/moca',
description:
'Every agent carries a verifiable identity, so the platform knows who is acting before it decides what they may do.',
'A dense, warm process-manager runtime for stateful AI coding agents — sessions resume exactly where they left off. Agent state is durable, tool execution runs in isolated sandboxes, and one supervisor serves many sessions per warm worker.',
},
{
number: '02',
status: 'Ready',
title: 'Authorization and access',
name: 'Cortex',
repo: 'rossoctl/cortex',
description:
'Agents run with least privilege and act by delegation, never with more authority than the user and the agent both hold.',
"See what your coding agent actually sends — and pay less for it. A proxy in the agent's request path that shows every model call, tool call, and cost as it happens, and strips the tool definitions the agent never calls. Cortex Local runs on individual, end-user machines; Cortex Remote, deployed along the harness, is being explored.",
},
{
number: '03',
status: 'beta in 0.7',
title: 'Intent-based access',
name: 'Eventing',
repo: 'rossoctl/eventing-lab',
description:
"Access is tied to what the user actually asked for, so an agent that drifts off task can't quietly reuse its access for something else.",
'The signed-event agent wake path: a Kafka message wakes a coding agent, and every step of its run streams back out as CloudEvents — with identity and signing on the event path.',
},
{
number: '04',
status: 'beta in 0.7',
title: 'Tool semantic validation',
name: 'AutoBench',
repo: 'rossoctl/autobench',
description:
"Tool calls are checked for correctness and meaning before they run, catching calls that are malformed or don't match the task.",
"Automated benchmarking of agentic AI workloads: deploys a benchmark workload — a tool-calling agent paired with the benchmark's MCP server — runs it, collects per-task telemetry, and publishes the results as durable artifacts.",
links: [{label: 'autobench-claudecode', href: 'https://github.com/rossoctl/autobench-claudecode'}],
},
{
number: '05',
status: 'alpha in 0.7',
title: 'Context compaction',
name: 'State management',
repo: 'rossoctl/state-lab',
description:
"The platform assembles and trims what goes into the model's context, keeping it relevant and lowering token cost.",
},
{
number: '06',
status: 'alpha in 0.7',
title: 'Data-flow analysis',
description:
'The platform tracks where the data behind a decision came from, so its provenance can gate what the agent does next.',
},
{
number: '07',
status: 'beta in 0.7',
title: 'Failure recovery',
description:
'The platform detects when an agent stalls or fails and returns it to a known state instead of leaving work half-finished.',
'Where agent state lives, who owns it, and what happens when a process dies mid-task — demo flows exploring durability and passability for distributed agents.',
},
];

export const contextGovernanceProjects = [
{
number: '08',
status: 'beta in 0.7',
title: 'User interaction',
name: 'Context Guru',
repo: 'rossoctl/context-guru',
description:
'For high-stakes steps, a person stays in the loop and approves before the agent acts.',
'Provider-agnostic context engineering for LLM agents — lower token cost and latency without sacrificing accuracy. Ships as plugins for Claude Code and Codex.',
links: [{label: 'project site', href: 'https://rossoctl.github.io/context-guru/'}],
},
];

export const rossoctlServices = [
{
number: '01',
title: 'Skills',
name: 'AIAC',
repo: 'rossoctl/aiac',
description:
'Reusable, governed capabilities an agent can draw on, versioned and controlled rather than pasted in.',
'AI-based access control: translates natural-language policy into enforced permissions on OPA, driven by Keycloak identities.',
},
{
number: '02',
title: 'Platform Tools',
name: 'Data governance',
repo: 'rossoctl/lab-data-governance',
description:
'Standalone tools the platform provides to strengthen agent and platform governance, security, resilience, reliability and efficiency, so each concern is handled once by the platform instead of rebuilt in every agent.',
'Data-aware governance and security: observes how data flows across agents, tools, and sessions, and derives interactions, payload classifications, and data lineage from the raw stream.',
},
{
number: '03',
title: 'Memory',
name: 'Context Service',
repo: 'rossoctl/context-service',
description:
"Durable memory that survives sessions and restarts, so an agent doesn't start from nothing each time.",
'Agent context infrastructure: durable storage and lifecycle management for agent state, memory, knowledge, workspaces, and artifacts — for local harnesses and container-native platforms alike.',
},
];

export const automationProjects = [
{
number: '04',
title: 'Knowledge base',
name: 'Automation',
repo: 'rossoctl/automation',
description:
'Shared, queryable knowledge agents use to ground their work in your own data.',
'Automated maintenance programs for the rossoctl org, built on a scanner/fixer pattern.',
},
{
number: '05',
title: 'Sandboxes',
name: 'RepoMan',
repo: 'rossoctl/repoman',
description:
'Isolated environments to run untrusted or self-modifying agents without risking the host or other tenants.',
'Automated maintenance programs — PR review, link health, dependency bumps, weekly reports — across GitHub repos you enroll once. Install once, enable programs on the fly.',
},
];

<div className="rosso-landing">

# Platform primitives for trustworthy AI agents
# Agent infrastructure, built in the open

<div className="rosso-intro">

rossoctl is a set of platform primitives for agent security, resilience, reliability and efficiency that
platform owners can build on. It comes in three parts: RossoCortex, a data plane that mediates
what agents do, a set of services agents can call and a set of tools and frameworks for observability,
security, governance and administration. It's open source, framework-neutral,
and built on open standards, supporting A2A and MCP.
rossoctl is an open source research project building the infrastructure AI coding agents need to run
for real: a dense runtime where sessions resume exactly where they left off, a proxy that shows what
your coding agent sends and what it costs, an event-driven wake path, automated benchmarking, and
state management for distributed agents. Around that core we work on context, access control, and
data governance, and we run the org itself on the automation we build.

</div>

<div className="margin-top--md margin-bottom--lg">
{/* Primary CTAs point at the project repos (docs hidden pre-launch). */}
<Link className="button button--lg cta-primary margin-right--sm" href="/docs/">Get started</Link>
<Link className="button button--lg cta-primary margin-right--sm" href="#core-projects">Get started</Link>

@aslom aslom Nov 2, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[low] The hero "Get started" now points at #core-projects, an on-page anchor,
while the adjacent GitHub button is a real external link. Both verified working on the
preview.

One consequence worth being deliberate about: /docs/ still exists and returns 200,
but nothing in the hero routes there anymore. That matches the PR's stated intent
(docs still present the earlier platform work), so no change requested — just
flagging that the docs are now reachable only via the navbar, which is easy to forget
when the docs section is eventually refreshed.

@mrsabath mrsabath Nov 2, 2026

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ack, and thanks for flagging the consequence. The remaining funnels into /docs/ (announcement bar, navbar Documentation + version dropdown) are being tracked as a follow-up until the docs themselves catch up with the current focus.

<Link className="button button--lg cta-secondary" href="https://github.com/rossoctl" target="_blank" rel="noopener noreferrer">GitHub</Link>
</div>

<ArchitectureDiagram />

## The problem
## Core projects

AI agents don't behave like the cloud applications we know how to run. They decide which tools to call while they run, they don't cleanly separate data from instructions, their goals drift as context changes, and they can't reliably report what they did. Those traits make the ordinary demands of production hard to meet: failure recovery, admission control, security, and resource management.
The projects we are actively building and shipping.

Agent systems fail in ways enterprise architecture wasn't built for, so they need a platform to supply the guarantees they can't supply themselves. Kubernetes solved the same kind of problem for cloud software by separating application logic from placement, scaling, and recovery. An agent platform needs an analogous split: decouple agent logic from the guarantees around admission, isolation, and failure recovery that production demands.
<ProjectCards items={coreProjects} />

## RossoCortex, the data plane
## Context &amp; governance

RossoCortex is an intercept. It sits transparently between an agent and everything external it touches: models, tools, users, and other agents. From that one vantage point it enforces the guarantees agents can't enforce for themselves. It is flexible by design: it works with different agent types, including black-box harnesses, integrates with different sandboxes, and works across container network choices.
Context, access control, and data governance for agent workloads.

<Accordion items={cortexFunctions} />
<ProjectCards items={contextGovernanceProjects} />

## Rossoctl services
## Automation &amp; repo maintenance

The services are the building blocks agents need to do real work: skills, tools, memory, knowledge bases, and sandboxes. The focus is on capabilities that lower the risk of agentic workloads and use resources more efficiently.
The automation that keeps the org and its repositories healthy.

<Accordion items={rossoctlServices} />
<ProjectCards items={automationProjects} />

</div>
Loading