assess-technical-debt
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted source code, creating an attack surface for indirect prompt injection where malicious content could influence analysis or result in cross-site scripting (XSS) in the generated HTML report.\n
- Ingestion points: Scans all source files in the project directory as defined in Step 1.\n
- Boundary markers: None. There are no delimiters or warnings to the sub-agents to ignore instructions embedded in the analyzed data.\n
- Capability inventory: File system read/write, sub-agent invocation, and execution of local CLI tools.\n
- Sanitization: Absent. Findings are aggregated and formatted into an interactive HTML report without explicit filtering or escaping of content extracted from source files.\n- [COMMAND_EXECUTION]: The skill executes multiple shell commands to manage its analysis environment and run external tools.\n
- Evidence: Uses git worktree, git status, git show, and git worktree list. It also suggests running CLI analyzers such as lizard, radon, gocyclo, and jscpd on the target codebase.\n- [REMOTE_CODE_EXECUTION]: The skill instructs sub-agents to perform dependency installations from the target codebase.\n
- Evidence: Step 2 states, 'Give that worktree its own dependency install rather than reaching the shared tree's install'.\n
- Risk: This could execute malicious scripts defined in build configurations (e.g., npm postinstall scripts). The skill mitigates this by requiring that these checks run in an isolated git worktree under $TMPDIR that is discarded after use.
Audit Metadata