inboundfy-base-formatador
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted Markdown artifacts from the local filesystem and has the capability to write the formatted results back to the disk. It lacks explicit data delimiters or instructions to ignore embedded commands, creating a vulnerability where an attacker could influence the agent via content in the input files.\n
- Ingestion points: Reads Markdown artifacts from paths specified in the
entradafield (SKILL.md).\n - Boundary markers: Absent. The prompt does not define delimiters to separate the input data from system instructions or notify the agent to ignore content inside the data.\n
- Capability inventory: The skill uses capabilities to read and overwrite files on the local filesystem to apply formatting changes.\n
- Sanitization: Absent. There is no logic provided to sanitize or validate the content of the Markdown files for potential injection attacks.\n- [NO_CODE]: The skill package consists purely of instructional Markdown and YAML configuration files; it does not include any executable scripts or binary files, significantly reducing the direct attack surface.
Audit Metadata