AIR says its fake AI skill passed scanner checks by using a mutable external link, exposing a blind spot in agent skill ...
AIR says static scanning failed to detect a skill that redirected to a controlled domain and later altered its payload.
Picture this scenario: An Anthropic Skill scanner runs a full analysis of a Skill pulled from ClawHub or skills.sh. Its markdown instructions are clean, and no prompt injection is detected. No shell ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results