earnings-signal-play
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process unstructured external data from public company filings and earnings transcripts to generate sales outreach.
- Ingestion points: Phase 1 (SEC filings) and Phase 2 (Earnings transcripts) in SKILL.md.
- Boundary markers: The prompt templates provided in references/framework-notes.md lack explicit delimiters or instructions to ignore instructions embedded within the source documents.
- Capability inventory: The agent is instructed to execute a local validation script (scripts/check-output.py) and generate outreach emails.
- Sanitization: No specific filtering or sanitization of the external financial text is performed before interpolation into prompts.
- [COMMAND_EXECUTION]: The skill includes and utilizes a local Python script (scripts/check-output.py) intended for validating that generated deliverables meet basic keyword and length requirements.
Audit Metadata