Agent Setup

Audit an agent harness for sandboxing, permissions, hooks and guardrail gaps

Audit Claude Code config for duplicate skills, shell hooks and read-only files

Browse and safely edit Claude Code and Codex settings, hooks, memory and rules

Inventory Claude Code hooks and flag those that post to the network

Install monitoring hooks for Claude Code and Codex and keep your own hooks intact

Route an AI coding agent through the Salesforce Models API with OrgJWT auth

Register local agent runtimes with CORAL and check that their logins still work

Create and maintain concise AGENTS.md and CLAUDE.md project instruction files

Audit a repo and suggest Claude Code settings.json permission rules

Start, list and manage cc-viewer projects running on a server

Install NVIDIA NemoClaw and its sandbox, tunnel and remote access on Linux

Set up and audit project config files, permissions, context docs and leaked secrets

Get set up with coding agents and Power BI with plain-language onboarding

Audit your agent setup for skills, memory, permissions, hooks and secrets hygiene

Look up which Wisp tool fits a task and what its runtime can reach

Create or update Codex hooks and feature flags for global or project scope

Add a Codex plugin from a GitHub repo to a personal or repo marketplace

Create and update custom Codex agent roles as standalone TOML config files

Handle permission requests from the local comfy agent for folders and hosts