agent-framework
Skill4 starsChanged 6 months agoMITView on GitHub
Triggers when the agent creates or modifies AI agents, registers tools, or changes fallback logic. Covers Microsoft Agent Framework SDK patterns, singleton client, and thread management.
In this repository5 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- adr-workflowSkill
- api-clientSkill
- databaseSkill
- releaseSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 2 are distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 9
Use singleton azure_openai_client from backend/app/azure_openai_client.py
- Very specific: line 10
Never instantiate AzureOpenAIClient() directly
- Very specific: line 14
All agents use AgentClientManager from backend/app/agent_client_setup.py
- Very specific: line 24
All agents MUST check azure_openai_client.is_configured()
- Distinctive: line 25
When False: use deterministic game logic, never call Azure
- Very specific: line 29
Create threads with agent_client_manager.create_thread()
- Distinctive: line 30
Use threads to maintain conversation context across interactions
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Microsoft Agent Framework Patterns
Client Access
- Use singleton
azure_openai_clientfrombackend/app/azure_openai_client.py - Never instantiate
AzureOpenAIClient()directly - Uses
DefaultAzureCredential(managed identity in Azure,az loginlocally)
Creating Agents
- All agents use
AgentClientManagerfrombackend/app/agent_client_setup.py - Create with
agent_client_manager.create_agent(name, instructions, model, tools, temperature) - Store agent references for reuse via module-level singletons
Tool Registration
- Use
agent_client_manager.create_function_tool(name, description, parameters, callable_fn) - Include clear descriptions and typed parameters
- Tools are registered at agent creation time
Fallback Mode
- All agents MUST check
azure_openai_client.is_configured() - When False: use deterministic game logic, never call Azure
- Wrap Azure SDK calls in try-except with context logging
Thread Management
- Create threads with
agent_client_manager.create_thread() - Use threads to maintain conversation context across interactions
- One thread per game session per agent
Architecture Decision
- See ADR-0018 for migration rationale
- SDK:
azure-ai-projects>=1.0.0,<2.0.0 - API version:
2025-05-01
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

