python-profiling
Warn
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions involve environment setup using pyenv and pip to download packages from official registries. The example workload is designed to fetch content from Wikipedia over HTTPS for profiling purposes.- [COMMAND_EXECUTION]: The documentation provides several shell commands for managing virtual environments, installing dependencies, and executing profiling scripts like ./timing.sh.- [PRIVILEGE_ESCALATION]: The skill mentions the use of sudo for DTrace commands, which requires elevated system permissions. While it includes safety warnings against improper use, it provides instructions that involve higher-privilege operations.- [INDIRECT_PROMPT_INJECTION]: The example workload processes external data from Wikipedia, creating an attack surface where malicious instructions could be ingested into the agent context.
- Ingestion points: External content fetched from Wikipedia via requests.get in program.py.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the fetched content as untrusted data.
- Capability inventory: The skill enables shell execution, environment modification, and the use of system-level profiling tools.
- Sanitization: No sanitization or validation of the fetched external content is described before the parsing and profiling phase.
Audit Metadata