Browse the directory

Expert Codex plugin capability pack for safe plugin scaffolding, manifest quality, MCP integration, and maintainable distribution.

Harden OpenClaw agent environments with secure defaults, policy boundaries, tool governance, and incident response playbooks.

Official AWS Labs MCP server for AWS Identity and Access Management that lets AI assistants inspect and manage IAM users, roles, groups, policies, and access keys, with policy simulation and an opt-in read-only mode.

Apache-2.0 Python framework for building visible, controllable, production AI agents and multi-agent services with event streaming, permission controls, workspaces, sandbox backends, middleware, MCP support, Mem0 memory, agent teams, and multi-tenant multi-session serving.

Commercial Cirra AI MCP server that connects Claude and other MCP clients to Salesforce orgs via OAuth for metadata, permissions, flows, Apex, and SOQL administration through natural language.

Expert MCP remote server trust review capability pack for auditing OAuth flows, transport security, tool permissions, data exfiltration risk, and vendor scope before connecting Claude Code to third-party MCP servers.

macOS app and bundled stdio MCP server that lets Claude access selected local services such as Calendar, Contacts, Messages, Location, Maps, Reminders, camera, microphone, screenshots, Shortcuts, utilities, and Weather.

MCP server for connecting Claude to MySQL databases with schema inspection, SQL query execution, SSH tunnel support, multi-database mode, and configurable write-operation controls.

A maintainer guide for securing agentic coding workflows in open-source repositories: version-controlled MCP configs, permission defaults, contributor PR review for AI-generated changes, hooks, sandbox boundaries, and public disclosure hygiene.

Source-backed agent that reviews active Claude Code sessions and configuration for security gaps, cross-references the official security guidance, and produces a ranked remediation plan covering permissions, MCP trust, prompt injection, credential handling, and hook safety.

Connect Claude to Algolia-managed MCP servers for curated public search and recommendation access or internal, user-scoped search and analytics workflows.

Source-backed agent that designs safe recurring Claude Code routines, turning a repeated task into a well-scoped, idempotent skill-driven workflow with clear triggers, permissions, and reporting, grounded in the official Claude Code skills docs.

Source-backed guide for scoping MCP servers, tools, permissions, hooks, and memory when building Claude Code subagents for focused review, research, and implementation workflows.

Source-backed agent that reviews the container isolation posture of a NanoClaw deployment, checking filesystem mounts, per-agent scoping, credential vault routing, messaging channel permissions, and scheduled tasks before agents run with real access.

A practical walkthrough of least-privilege permission design in the Claude Agent SDK. Learn the evaluation order (hooks → deny → mode → allow → canUseTool), how allowedTools and disallowedTools differ, the six permission modes, and how to lock an agent to a fixed tool surface with dontAsk.

MCP Skills integration specialist for remote server configuration, tool permissions, multi-MCP orchestration, and Claude Desktop ecosystem workflows.

Expert capability pack for reviewing Claude Code autoMode settings blocks, trusted infrastructure prose, classifier rule overrides, and documented claude auto-mode CLI inspection before enabling permission-free auto mode.

Community reusable agent prompt for reviewing new MCP server adoption in Claude Code using official security documentation: trusted providers, permissions configuration, trust verification, and settings checked into source control.

Connect Claude to BrowserStack for permission-scoped web, app, accessibility, and test automation workflows.

Configure Claude Code auto mode hard-deny rules that block high-risk actions unconditionally, complement soft-deny prompts and team permission policy.

A reusable agent prompt that reviews a Claude Code auto mode setup: whether autoMode.environment trusts the right repos, buckets, and domains, how the allow, soft_deny, and hard_deny tiers resolve, and whether $defaults is kept across settings scopes.

A practical walkthrough of Claude Code permission modes for teams: what each mode allows, how to switch and default them, protected paths, and how administrators enforce or lock modes with managed settings.

A source-backed safety checklist for introducing Claude Code hooks with scoped events, minimal authority, reviewable commands, clear ownership, and a rollback path.

Build autonomous agents with the Claude Agent SDK and Claude Code subagents: the query loop, built-in tools, subagent delegation, and permission controls.

Microsoft-maintained MCP server and security-oriented WebAssembly component runtime for loading sandboxed tools, managing component permissions, and exposing Wasm component tools to Claude through MCP.

Set up Claude Code GitHub Actions for pull request review: install the Claude GitHub app, store ANTHROPIC_API_KEY in secrets, pin anthropics/claude-code-action to an audited commit SHA with least-privilege permissions, and follow documented security practices.

Roll out Claude Code server-managed settings for Teams and Enterprise: configure JSON in Claude.ai Admin Settings, understand delivery and caching, verify with /permissions, and use forceRemoteSettingsRefresh when fail-closed startup is required.

Production-oriented quickstart for the Claude Agent SDK: project bootstrap with agent-sdk-dev plugin, query loops, permissions, observability hooks, and deployment guardrails.

Guide to delegating engineering work from Slack to Claude Code: channel setup, task contracts, permissions, status updates, and safe escalation paths.

Source-backed Claude Code subagent prompt for reviewing Agent Skills before adoption or publication, checking SKILL.md scope, descriptions, invocation control, supporting files, tool permissions, helpfulness, safety, and privacy risks against official Claude Code skills guidance.