A runtime spend circuit breaker for agent trajectories: per-run budgets, fan-out and wall-clock caps, a graceful degradation ladder, and the cost attribution the OpenTelemetry GenAI conventions leave out.

agentsaicost-optimizationfinopsllm

cloudandsre-python

Repo · 0 stars · Python

▲ 0

Shared Python utilities for AI-native cloud and SRE tools — retry/throttle/circuit-breaker decorators, prompt-cache helper, token-cost meter. Used by alert-explainer, incident-scribe, and the rest of the cloudandsre.com toolkit.

aicircuit-breakercost-trackingllm

python

mcp-chaos

Repo · 0 stars · Python

▲ 0

Chaos engineering for the MCP tool-call plane: fault injection + reliability SLIs for agent tool use

chaos-engineeringfault-injection

mcpagentic-infrastructureai-agents

runbook-rag-starter

Repo · 0 stars · Python

▲ 0

The smallest useful RAG, in 40 lines — a Markdown folder, sqlite-vec, and a CLI. Companion to the 'What is RAG?' primer at cloudandsre.com.

aiembeddingsllm

pythonrag

sre-ai-toolkit

Repo · 0 stars · Python

▲ 0

Curated AI-assisted scripts for platform & reliability engineers

aiautomationcloudllmplatform-engineering

sre-harness

Repo · 1 stars · Python

▲ 0

A minimal, runnable reference harness for SRE agents: tool orchestration, verification, memory, guardrails, and observability in one ~40-line loop. Agent = Model + Harness.

aiai-agentsguardrailsharness-engineeringllm