constant-time-analysis
Fail
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- COMMAND_EXECUTION (HIGH): The skill instructs the agent to run an unprovided and unreviewed script using
uv run {baseDir}/ct_analyzer/analyzer.py. This allows arbitrary code execution from within the skill's package directory. - COMMAND_EXECUTION (HIGH): The documentation in
references/php.mdprovides commands for the agent to usesudofor system-level installations and file operations (e.g.,sudo make install,sudo cp). - EXTERNAL_DOWNLOADS (HIGH): The skill instructs the agent to download and install third-party code from non-whitelisted domains such as
pecl.php.netviacurlandpecl, andgithub.com/derickr/vld.gitviagit clone(the organizationderickris not on the trusted list). - COMMAND_EXECUTION (HIGH): The
SKILL.mdfile contains instructions for the agent to modify the user's shell profile (~/.zshrc) to add to the PATH, which is a persistence-related modification. - PROMPT_INJECTION (LOW): The skill ingests and processes untrusted user-provided source code. This creates a surface for indirect prompt injection where malicious code comments or metadata could attempt to influence the agent's analysis or subsequent actions.
Recommendations
- AI detected serious security threats
Audit Metadata