Orbit Support Agent — AI skill for Claude Code
A small, runnable demo of the one idea that changed in Pydantic AI 2.0: you stop.
How to install Orbit Support Agent
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open coleam00/orbit-support-agent and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Orbit Support Agent does
A small, runnable demo of the one idea that changed in Pydantic AI 2.0: you stop.
Alternatives in AI
- Anthropic Quickstarts — by Anthropic - Offers comprehensive development guides for three distinct AI-powered demo projects with standa 15.4k ★
- Upstream Watch — Review pydantic-ai and pydantic-ai-harness releases published since the last watch entry, and record a verdict 361 ★
- Execute Pydantic AI Prp — Execute Pydantic AI Agent PRP 132 ★
README
Orbit Support Agent — the Pydantic AI 2.0 capability pattern
A small, runnable demo of the one idea that changed in Pydantic AI 2.0: you stop **configuring** an agent through a giant constructor and start **composing** it out of *capabilities*.
The idea
A **capability** is one composable brick that bundles everything that used to be scattered across the `Agent(...)` call — its instructions, its tools, its model settings, and its lifecycle hooks — into a single self-contained unit. You build an agent by snapping bricks together, and the same brick drops into a completely different agent, unchanged.
from pydantic_ai import Agent
from pydantic_ai.capabilities import Thinking
agent = Agent(
"openrouter:anthropic/claude-sonnet-4.6",
capabilities=[
KnowledgeBaseCapability(), # its instructions + its tool + model settings + a guardrail
EscalationCapability(), # its own tool + its own instructions
Thinking(effort="low"), # a built-in capability, one line
audit_hooks(), # lifecycle hooks, no subclass needed
],
)
That is the whole thesis: **compose, don't configure.** Each brick carries its own instructions, tools, settings, and hooks, so nothing has to be hand-wired onto the agent and nothing is welded to one model. The payoff is reuse — pick a brick up and drop it into the next agent without touching it.
What's in here
| File | What it shows |
|---|---|
cli.py |
An interactive REPL. After every answer it prints a "capabilities used this turn" panel that maps each tool call back to the capability that owns it. /faq swaps to an agent that reuses the KnowledgeBase brick, live. |
main.py |
The agent composed from capabilities, then the reuse payoff: the same brick dropped into a second, unrelated agent without a single change. |
capabilities.py |
The capabilities themselves — each one a small class bundling its own instructions, tool, settings, and ho |
Related Skills
Personal Digital Brain With AI
An agentic Obsidian vault, split into six small repos that each show one idea: layout, privacy guard, frontmat
Orbit Skills
Official Claude Agent Skills for Orbit, the AI-first Logistics Operating System.
AI Monitor P4 S3
LVGL AI desk companion for GUITION ESP32-P4 and experimental Waveshare ESP32-S3 4.3-inch touchscreens. NOVA/OR
AI Projects Hub
20 real-world AI projects with runnable code — RAG, agents, automation, and India-first tools using Python and
Claude Code Source All In One
Explore Claude Code source architecture and learn AI agent internals with runnable code, deep dives, and educa
Pagewell
Agent skill + CLI for PageWell: write a designed document or a runnable single-page artifact with your AI agen
Related Agents
30 Agents Every AI Engineer Must Build
Agent-architecture books usually stop at diagrams, so this Packt companion repository (by Imran Ahmad, 2026) s
Value Critic
Upstream VALUE gate. Judges ONE candidate idea BEFORE any code is written — does it plausibly move the product
Scaffold Agent
Projects the approved GDD into a runnable Phaser 4 skeleton and machine-readable implementation contracts.