scope-creep-detector
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The script
scripts/scope_creep.pyexecutes localgitcommands using thesubprocess.runfunction. While it uses argument lists to mitigate shell injection, it relies on the presence of a local git installation and performs operations based on user-provided repository paths and branch names. - Evidence: The
run_gitfunction inscripts/scope_creep.pyexecutes commands likegit diffandgit branchon paths provided via the--repoargument. - [INDIRECT_PROMPT_INJECTION]: The skill processes
git diffoutput, which constitutes untrusted external data. A malicious diff could contain embedded instructions intended to influence the agent's behavior when it reviews the analysis report. - Ingestion points:
scripts/scope_creep.pyreads diff data from the local repository or provided input files. - Boundary markers: Absent. The script parses the diff but does not provide specific delimiters or warnings to prevent the LLM from following instructions embedded within the hunks.
- Capability inventory:
scripts/scope_creep.pyperforms local command execution and provides structured analysis to the agent. - Sanitization: Absent. The script does not sanitize or filter natural language instructions from the diff content before it is presented to the agent.
Audit Metadata