streak
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill contains instructions to fetch external data through a 'Research' feature (Flow 2, Step 3.5 in references/flows-detailed.md). This feature searches for news and updates related to the user's challenge topic on the public internet.
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted content from the web during the research phase.
- Ingestion points: External search results fetched from the web in references/flows-detailed.md and references/types.md.
- Boundary markers: Absent; there are no instructions to use delimiters or ignore instructions within the fetched research data.
- Capability inventory: The skill has the ability to read and write files and execute shell commands to manage the .streak/ directory structure.
- Sanitization: The instructions do not include any steps for sanitizing or escaping the content retrieved from the web before it is presented to or processed by the agent.
- [COMMAND_EXECUTION]: The skill uses shell commands, such as mkdir -p, to dynamically create and manage the directory structure for sessions and challenges within the .streak/ folder.
Audit Metadata