remove-ai-slops
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to measure file size using a shell command:
awk '!/^[[:space:]]*$/ && !/^[[:space:]]*(#|\/\/)/' <file> | wc -l. The<file>variable is interpolated directly into the shell command. If a file path in the branch diff or the provided arguments contains shell metacharacters (e.g.,;,|, or backticks), it could lead to arbitrary command execution in the shell environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes source code from branch diffs or external file lists. This code is passed to sub-agents for analysis and cleanup. \n - Ingestion points: Source files identified via
git diffor passed as arguments inSKILL.md. \n - Boundary markers: The instructions to the sub-agent do not include explicit delimiters or instructions to ignore embedded prompts within the source code content. \n
- Capability inventory: The agent can execute shell commands (
git,awk), project test runners, and fileEditoperations. \n - Sanitization: The skill does not perform sanitization on the file names or the content of the files being processed, relying instead on regression tests to catch behavior changes.\n- [DYNAMIC_EXECUTION]: The skill relies on executing project-specific external tools (test runners, linters, type-checkers, and security scanners) that are determined at runtime. While necessary for the skill's functionality, this behavior introduces a dependency on the integrity of the local environment's tooling.
Audit Metadata