claude-flow
Verified · 3 days agoAI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
agent-firewall-mcp
Verified · 18 days agoInput/output safety for AI agents: known-pattern injection and obfuscation scan, URL/IP, secrets.
Install
claude mcp add agent-firewall-mcp -- npx -y [email protected]
Our take
This server bundles several agent-safety checks such as prompt-injection screening, URL and IP validation, password breach checks, and PII scanning. It is best suited to developers putting guardrails around agent inputs and outputs, especially in enterprise or security-conscious settings. The scope is practical, but the current maturity signal is modest, so it should be treated as a useful layer in a defense-in-depth setup rather than a complete security boundary.
Verification record
- 18 days ago
- 0
- 18 days ago
- no
- MIT
- 326
- 1.2.0
- 2026-07-01
More in ai
browse all →tldraw
Verified · 5 days agoDraw and visually collaborate with your agents on tldraw's canvas.
Codebase Memory
Verified · 22 days agoCodebase knowledge graph for AI agents — 159 languages, sub-ms queries, 99% fewer tokens.
OpenMetadata
Verified · 17 days agoOfficial OpenMetadata MCP: governed context and business semantics for AI assistants and agents.
mobilerun
Verified · 6 days agoControl real Android and iOS devices with LLM agents — tap, swipe, type, automate flows.
PraisonAI
Verified · 21 days agoAI Agents Framework with Self Reflection and MCP support