search-docs

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The prompt explicitly instructs the subagent to run with mode "bypassPermissions" (i.e., to circumvent access controls) and to return results verbatim, which are deceptive/unauthorized instructions outside a normal documentation search.

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt instructs calling a docs-searcher with "bypassPermissions" and to report results "exactly as returned," which forces the agent to return any sensitive secrets or credentials present in the documentation verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill explicitly instructs using a "docs-searcher" subagent with mode "bypassPermissions", which requests evasion of access controls and therefore indicates intentional unauthorized access and high risk of abuse.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The prompt explicitly instructs calling a subagent with mode "bypassPermissions", which directs the agent to bypass security controls and therefore compromises the host environment.