bridge-mind/BridgeWard

Trust nothing. Ship safely. — Skeptical-reading and prompt-injection defense skill for AI agents. Provenance tagging, red-flag patterns, refusal templates, and a read-only injection auditor. MIT.

38Stars on the repository
4Mods indexed here, across every type
4mo agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

injection-auditor

01

bridge-mind/BridgeWard

Agent

Read-only auditor that scans files, directories, URLs, and MCP tool descriptions for prompt-injection attempts. Reports hidden text, override phrases, exfil constructs, fake structural markers, repo-poisoning artifacts, and rug-pull MCP descriptions with severity-tagged findings and remediation suggestions. Use for…

38 4mo ago C 78 tokens original MIT