inboundfy-copy-edicao
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and transform external text content provided by the user, creating a potential surface for indirect prompt injection if the source text contains adversarial instructions.
- Ingestion points: The skill receives a file path and processes the 'texto recebido' (received text) as described in the
Fluxosection ofSKILL.md. - Boundary markers: The instructions mandate the use of
inboundfy-anti-slopat checkpoints A0 and A5 to distinguish between the input signals and the revision changes, effectively acting as a logical delimiter. - Capability inventory: The skill possesses the capability to read local project configuration files (e.g.,
.inboundfy/context/empresa.md) and generate revised files and reports. - Sanitization: The
inboundfy-anti-sloptool is used as a mandatory validation layer to ensure the output remains focused on the primary task and is not diverted by instructions embedded within the input copy.
Audit Metadata