scrollclaw-score
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from campaign briefs and research files which could be exploited if those files contain malicious instructions.\n
- Ingestion points:
workspace/campaigns/<slug>/brief.mdandworkspace/campaigns/<slug>/persona-research.mdare used as context for the virality scoring process.\n - Boundary markers: Absent; there are no instructions to the agent to ignore embedded commands within these files.\n
- Capability inventory: The skill has the ability to read and write files within the workspace and execute shell commands.\n
- Sanitization: Absent; no validation or escaping of external content is described.\n- [COMMAND_EXECUTION]: The skill instructions provide a shell command example (
bash scripts/batch-campaign.sh) for batch processing, which relies on the integrity of scripts within the project repository.
Audit Metadata