Mend is an AI agent that repairs Playwright E2E tests broken by selector drift. It investigates the live DOM via tool calling, proposes a fix, and only accepts it after re-executing the test — never on the model's word alone. Low-confidence fixes route to human review instead of auto-merging.
End-to-end retail analytics consulting project built to BCG X standards churn prediction, price elasticity, marketing mix modelling, SHAP explainability, hypothesis testing, and an AI Copilot (agentic tool-use loop) all production Python, no notebooks.
Tool Governance for Pydantic AI Agents
A minimal, retrofittable injection-to-RCE containment loop for tool-using AI agents that accept untrusted file attachments
A Model Context Protocol (MCP) server in Go that provides a splunk_search tool, enabling LLMs to query Splunk instances via the REST API
Autonomous diagnostic agent running natively in a Salesforce org — Apex, LWC, Claude API with native tool_use, live Platform Event streaming.
File-watcher daemon that turns messy markdown meeting notes into clean, well-structured Linear issues. Anthropic forced tool use with a Pydantic-mirrored schema, idempotent batches via sidecar markers, and an eval harness with a no-op chitchat fixture as the hallucination regression catch. Built with the gstack workflow for Kalpi.
To associate your repository with the llm-tool-use topic, visit your repo's landing page and select "manage topics."