LaunchKit · 2026
Back to Skills

symbiont

**Purpose**: This guide helps AI assistants quickly build secure, compliant Symbiont agents following best practices.

0
454 downloads
by @jaschadub

Setup & Installation

openclaw skills install @jaschadub/symbiont

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install symbiont

Version History

v1.0.2Feb 27, 2026 - Version note by skill developer:

Key updates from v1.4.0 → v1.5.0: - Description: Added ORGA reasoning loop, Cedar policy authorization, knowledge bridge - What Makes Symbiont Unique: Added 4 new bullet points (ORGA, Cedar, Knowledge Bridge, Durable Journal); updated crypto verification to mention AgentPin - New section: Agentic Reasoning Loop — covers minimal loop setup, typestate phase transitions, all 7 journal event types, Cedar policy gate with entity type mapping, and knowledge bridge integration - Documentation links: Added reasoning-loop.md and security-model.md, removed stale tool_review_workflow.md link

What This Skill Does

Symbiont is an agent runtime for building secure, policy-enforced AI agents using a declarative DSL. It implements a typestate-enforced ORGA reasoning loop (Observe-Reason-Gate-Act), Cedar-based authorization, and multi-tier sandboxing via Docker, gVisor, and Firecracker. Compliance patterns for HIPAA, SOC2, and GDPR are built in.

Policy enforcement happens at the framework level via Cedar authorization and compile-time phase transitions, so security rules cannot be bypassed by mistakes in agent logic.

When to use it

  • Processing GitHub webhook events with HMAC signature verification
  • Running scheduled data archival or cleanup agents on cron schedules
  • Orchestrating multi-step workflows with failure recovery and circuit breakers
  • Building parallel API aggregators that call multiple external services
  • Scanning codebases for vulnerabilities inside isolated gVisor sandboxes

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: Build a webhook processor that validates GitHub PR event signatures and routes opened pull request events to a code review agent

AGENT
  1. 1Define a webhook block with the GitHub provider preset and a vault-referenced HMAC secret
  2. 2Write a webhook_processor agent with a policy allowing JSON parsing but denying code execution and file access
  3. 3Verify the X-Hub-Signature-256 header using the verify_hmac_sha256 built-in before touching the payload
  4. 4Configure a filter on $.action == 'opened' so only relevant events reach the agent
  5. 5Set memory to ephemeral and timeout to 5000ms, returning HTTP 200 on success or 401 on invalid signature
OUTPUT

A deployed /hooks/github endpoint that validates GitHub signatures, filters for opened PR events, and routes them to the code review agent while rejecting all others with HTTP 401