Skip to content
openagentix openagentix docs

Roadmap

0.1 – Core (in progress)

Section titled “0.1 – Core (in progress)”
  • Control node: API (Fastify, OpenAPI 3.1), PostgreSQL, run queue, console (React), OIDC/LDAP sign-in, RBAC with six roles, API tokens.
  • Agents as code: agents.md parser and validator, immutable published versions, pipelines of 1..n agents.
  • Events: signed webhooks, Kafka, cron, mail-in via webhook; Teams and Slack through generic webhooks.
  • Guardrails: per-agent audit gate, global control agent, approvals, data classification.
  • Audit: append-only SHA-256 hash chain with Ed25519-signed checkpoints and verification.
  • Costs: per step and run, price table, budgets with hard stop.
  • Providers: OpenAI-compatible, Ollama, AWS Bedrock (VPC endpoint, proxy, IRSA), Anthropic, simulated.
  • MCP: client with per-agent allowlists, timeouts and result limits; the policy gate as an MCP proxy.
  • Runners: in-process and local (oax CLI), with the interfaces for all later runners.
  • Observability: Prometheus metrics, OpenTelemetry traces, JSON logs, health endpoints.
  • Tenants and access: tenants as the isolation boundary, roles per tenant and optionally per agent (details).
  • Bring your own: provider keys by reference and a model catalog from a pinned models.dev snapshot (details); MCP servers per tenant.
  • Cost attribution: per tenant, agent, use case and run, with CSV/JSON export.
  • Delivery: Docker Compose, Helm chart for Kubernetes and EKS, this website and docs.

0.2 – Worker nodes

Section titled “0.2 – Worker nodes”
  • container runner (Docker, Podman) and kubernetes-job runner (also EKS) that spawn one short-lived worker per run.
  • Toolbox images from a catalogue: minimal, non-root, read-only, pinned by digest, signed (cosign), with SBOM (syft) and vulnerability scans (trivy); egress allowlisted per toolbox.
  • Per-run secret injection and revocation.

0.3 – Anywhere

Section titled “0.3 – Anywhere”
  • aws-lambda, github-actions and gitlab-ci runners with signed callbacks.
  • Optional external harnesses: Claude Code, OpenCode, Hermes, OpenClaw, all routed through the policy gate.
  • Dedicated Teams and Slack adapters, IMAP polling for mail.

Later: planning, governance, factory

Section titled “Later: planning, governance, factory”
  • Agent Check and Agent Build: from a process described in plain language to a least-privilege agent plan and a scaffolded, tested agent.
  • Governance: approval workflows, an MCP catalog, emergency security overrides.
  • Change-gated schedules (details).
  • Development guidelines and a global hardening agent (details).
  • Dark software factory, opt-in and recommended for proof-of-concept work only (details).
  • Air-gapped operation documented end to end (details).

The order within this list is not fixed; the platform repository tracks the current plan.

Towards 1.0

Section titled “Towards 1.0”
  • Stable v1 APIs and agents.md schema (openagentix.io/v1).
  • Signed releases and images, documented upgrade paths.
  • Chart repository and OCI publishing, public demo at demo.openagentix.si.

Have an idea? Open an issue in the platform repository; larger changes start as an ADR.