recoup-internal-consulting-content-drafter
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection by ingesting and processing data from external sources.
- Ingestion points: Reads signal data from the
signals/directory and extracts context from research wiki pages viaintegrations/research/AGENTS.mdin Step 2. - Boundary markers: The instructions do not define clear boundaries or delimiters to separate the external research data from the agent's task instructions, nor do they instruct the agent to ignore potentially malicious content within those sources.
- Capability inventory: The skill enables writing generated content to the workspace file system under
content/03-drafts/and involves chaining with other internal skills belonging to the same vendor, such asrecoup-internal-consulting-copy-editor,recoup-internal-consulting-article-illustrator, andrecoup-internal-consulting-graphics. - Sanitization: No evidence of content sanitization, validation, or filtering is specified for the ingested wiki or signal data before it is incorporated into the draft output.
Audit Metadata