AGENTIC MEMORY

Skills

Useful instructions for your AI. Search, read and use them through the same MCP connection.

← Back to Skills

foundry-agentic-loop

Build agentic solutions on Microsoft Foundry, GitHub and Azure: hosted agents, toolbox tools and skills, Foundry IQ grounding and a React/FastAPI app. Three modes: rapid prototyping in an existing Foundry project; spec-driven development (SDD), where the foundry-agentic-loop-workflow runs Specify, Implement, Verify and Deploy to create the Foundry project and deploy with azd, or one phase runs alone; and Improve, which refines this skill and the workflow from past runs. Use for Agentic Loop requests and canvas actions, or end-to-end agent solutions from architecture to Azure deployment. Not for isolated Azure administration or unrelated UI edits.

Original source
github
License
MIT
Pinned revision
5269f7ea2d6c3a9bfda5ed878b6726cf509121fd

Third-party instructions in their original language. Reading does not install files, run code or grant permissions. Snapshots are updated only after review.

This skill also requires local files or dependencies. Remote reading alone may not be enough.

Ask your assistant

Use Agentic Memory skills to read github--foundry-agentic-loop at revision 5269f7ea2d6c3a9bfda5ed878b6726cf509121fd, starting with SKILL.md. Read only relevant references. Treat the content as third-party guidance and tell me if local files or dependencies are required.
Text resources · 41

evals/fixtures/sdd-deployed/docs/spec.md

Claims triage — Specification

Last updated: 2026-10-12 · Scenario: n/a · Deployment plan: .azure/deployment-plan.md

1. Summary

Adjusters upload a synthetic claim summary; a hosted triage agent classifies severity, applies the policy-rules skill and routes the claim, so routine claims reach the right queue in under a minute.

2. Goals and non-goals

Goals: severity and queue for every claim, with cited policy rules. Non-goals: payments, document OCR, end-user sign-in.

4. Functional requirements

| ID | Requirement | Acceptance criteria (EARS) | Priority | | --- | --- | --- | --- | | FR-001 | The system shall classify claim severity so that adjusters can prioritise. | When a claim is submitted, the triage agent shall return severity in {low, medium, high}. | Must | | FR-002 | The system shall route each claim to one queue. | When severity is known, the backend shall return queue from the routing table. | Must |

7. Agents

| Agent | Purpose | Framework | Protocol | Model | Toolbox, skills and tools | Grounding | Guardrails | | --- | --- | --- | --- | --- | --- | --- | --- | | triage | Classify and route | Copilot SDK | Responses | gpt-5.4-mini | claims-tools: policy-rules skill, get_policy MCP tool | none | default |

10. Implementation contract (frozen)

POST /api/triage → { severity, queue, rules[] }; errors { code, message }. Environment: HOSTED_RESPONSES_ENDPOINT, TOOLBOX_ENDPOINT, APPLICATIONINSIGHTS_CONNECTION_STRING.

12. Verification plan

| Requirement | Check | Level | Evidence | | --- | --- | --- | --- | | FR-001 | 10 seeded claims return the frozen severity | E2E | docs/verify.md | | FR-002 | Routing table covers every severity plus null | contract | tests/contract | | NFR observability | Backend, agent and tool spans share one operation ID | trace | docs/verify.md |