pandas-data-manipulation-rules
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill contains a 'Memory Protocol' that instructs the agent to run a shell command (
cat .claude/context/memory/learnings.md) at the start and end of a session. This is a mechanism for maintaining learning state across resets by reading and writing to a specific local file path. - [PROMPT_INJECTION]: The skill operates on user-controlled Python files (matching
**/*.py), which introduces an indirect prompt injection surface where malicious code comments or strings could attempt to influence the agent's behavior. - Ingestion points: Source code files identified by the
**/*.pyglob pattern inSKILL.md. - Boundary markers: The instructions do not specify the use of delimiters or clear separators to distinguish between the agent's instructions and the untrusted code being reviewed.
- Capability inventory: The skill is configured with
Read,Write, andEdittools as seen in the YAML frontmatter ofSKILL.md. - Sanitization: There is no evidence of input validation, escaping, or instruction-override protection for the data read from the user's codebase.
Audit Metadata