inboundfy-base-validador
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to audit external content such as 'asset candidate', 'brief', and 'materials-source'. These inputs are ingested into the agent's context during the processing flow without specified boundary markers or sanitization, creating a risk that malicious instructions embedded within these files could influence the agent's validation logic.\n
- Ingestion points:
SKILL.md(Fluxo Steps 2 and 3) reads external asset, brief, and source material files into the context.\n - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill definition.\n
- Capability inventory: The skill calls other tools (
inboundfy-copy-editor,inboundfy-anti-slop) and executes a shell command (inboundfy content digest).\n - Sanitization: The instructions do not describe any sanitization or validation of the text content before it is processed by the agent.\n- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to manage the environment and verify file integrity.\n
- Evidence:
SKILL.md(Fluxo Step 9) directs the agent to executeinboundfy content digest <id>. The<id>parameter is derived from the input and could potentially be used for command injection if the underlying platform does not provide adequate sanitization.\n - Evidence:
SKILL.md(Verificação do setup) provides instructions for the user to executeinboundfy-setupif the local environment is missing required framework files.
Audit Metadata