recoup-internal-consulting-friday-review
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data from various sources which could contain hidden or adversarial instructions intended to influence the agent's behavior during the review process.
- Ingestion points: Reads raw transcripts from
content/01-raw/, dated signals fromsignals/, and files inknowledge/andintegrations/. - Boundary markers: None identified in the instructions to delimit external content or warn the agent about embedded instructions.
- Capability inventory: Invokes several internal tools (
recoup-internal-consulting-*) for content recycling, extraction, research mining, and drafting. - Sanitization: No explicit sanitization or validation of the ingested content is mentioned.
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute multiple other capabilities/tools by name to process data and generate reports.
- Evidence: Execution of
recoup-internal-consulting-content-recycler,recoup-internal-consulting-content-extraction,recoup-internal-consulting-research-miner,recoup-internal-consulting-content-idea-generator, andrecoup-internal-consulting-content-drafter.
Audit Metadata