doomscroll
Warn
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: MEDIUMDYNAMIC_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill provides the full source code for a Python script (
doomscroll.py) and instructs the agent to save it locally and execute it usingpython3to merge and rank results. - [COMMAND_EXECUTION]: The agent is instructed to perform multiple shell command executions, including running the
tregCLI for data collection and the generated Python script for data processing. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from various social media platforms (X, TikTok, YouTube, Reddit, Hacker News, Product Hunt).
- Ingestion points: Data is fetched from public social media feeds and comments via external tools (SKILL.md).
- Boundary markers: Absent. The agent is instructed to read the raw content and comments to "judge" the stories without explicit instructions to ignore embedded commands.
- Capability inventory: The agent has the capability to write files to the local system and execute shell commands (SKILL.md).
- Sanitization: No sanitization or instruction-filtering is performed on the ingested social media content before the agent processes it.
Audit Metadata