kashifsayyad09/incident-pilot-agent
Developer toolsЭта страница переведена автоматически. Технические детали смотрите в оригинале. Открыть оригинал
IncidentPilot AI is an intelligent AWS incident troubleshooting agent powered by Strands Agents, Strands Decider 2B, and Amazon Bedrock.
Обзор
Read-only AWS incident investigation agent for the (category: Tool Selection, #strands-decider-2b #tool-selection). Incident → Strands Decider 2B picks the next diagnostic tool → read-only AWS diagnostic → Amazon Bedrock analysis → incident report → dashboard. Nothing in the UI is faked: unavailable components are shown as unavailable. Requirements: Python 3.11+ (tested 3.13), Node 20.9+ (tested 22). Or with Docker: docker compose up --build. - No credentials in the repo; use IAM roles / SSO. .env is git-ignored. - AWS access is read-only by design. Wildcard CORS is rejected. - Set INCIDENTPILOT_API_KEY to require X-API-Key on /api/v1/*.
README
Показан фрагмент оригинального README. Открыть оригинал.
IncidentPilot AI
From Cloud Chaos to Clarity.
Read-only AWS incident investigation agent for the AWS Strands Decider Agent
Hackathon (category: Tool Selection, #strands-decider-2b #tool-selection).
Pipeline: Incident → Strands Decider 2B picks the next diagnostic tool → read-only AWS diagnostic → Amazon Bedrock analysis → incident report → dashboard.
Status: Phase 3 (Bedrock analysis agent) complete, pending live Decider/Bedrock verification
| Area | State |
|---|---|
| FastAPI backend skeleton, config, schemas, API-key option, CORS | Done, tested |
| Next.js frontend: 5 routes, sidebar app shell, R3F 3D background, GSAP motion, mode badge, API client | Done, builds, screenshot-checked in headless Chromium |
| Simulation catalogue (5 scenarios, metadata only) | Done |
Decider 2B tool selection (backend/decision_patterns/tool_selection/) |
Implemented against the real strands-decider System One API; confidence-gated; reusable pattern with guide. Not yet run against the real model (see below) |
Bedrock / Strands agent (backend/app/agent/) |
Implemented: structured analysis, evidence-id grounding, report store, /api/v1/investigations. Tested with a Bedrock reply test double and the real offline path; no real Bedrock call has been made (see below) |
| AWS diagnostics (CloudWatch, ALB, EC2, RDS) | Not implemented (Phase 4) |
| Live metrics | Not implemented (need Phase 4); shown as “—” / unavailable |
| Reports (history, detail, JSON/Markdown export) | Done for simulation incidents |
Nothing in the UI is faked: unavailable components are shown as unavailable.
Workflow Diagram
Amazon Bedrock: setup and verification status
aws bedrock list-inference-profiles --region us-east-1 # pick one enabled in your account
# .env: INCIDENTPILOT_BEDROCK_MODEL_ID= AWS_REGION= (IAM role / SSO, no keys in files)
cd backend && pytest -m live_bedrock -s # one real call; skipped if not configured
Without a model id or credentials /api/v1/investigations returns 503; no report is ever
fabricated. Reports are labelled AI-generated and advisory; confirmed facts are checked
against numbered evidence and demoted to hypotheses when unsupported.
Verified: Strands structured-output wiring against a stubbed Bedrock reply, the real no-credentials failure path, error mapping, grounding, the API and UI. Not verified: a real Bedrock call (no AWS credentials in the development sandbox).
Decider 2B: setup and verification status
pip install strands-decider # pulls torch; downloads StrandsAgents/strands-decider-2B-hobson-v19 on first run
strands-decider serve StrandsAgents/strands-decider-2B-hobson-v19 --port 9000
# then set INCIDENTPILOT_DECIDER_URL=http://localhost:9000 in .envРекомендуемые инструменты
Попробуйте другой запрос или уберите фильтр.
Установка
npx skillfish add kashifsayyad09/incident-pilot-agent