legacy-modernizer
Fail
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: CRITICALEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONMETADATA_POISONINGINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [METADATA_POISONING]: Automated security scanners have flagged the instruction file 'SKILL.md' as malware (FileRepMalware). Additionally, there is a discrepancy between the author identified in the skill metadata (Jeffallan) and the platform-provided author (emmraan), which suggests potential impersonation or unauthorized distribution.
- [EXTERNAL_DOWNLOADS]: The skill provides a documentation link to 'https://jeffallan.github.io/claude-skills/skills/specialized/legacy-modernizer/' which is confirmed to be on a malicious URL blacklist by reputation scanners.
- [COMMAND_EXECUTION]: In 'references/system-assessment.md', the 'identify_hotspots' function utilizes 'subprocess.run' to execute 'git' shell commands on the host environment. This allows the agent to perform unauthorized system operations under the guise of technical assessment.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external codebases as its primary ingestion point (identified in 'references/system-assessment.md'). It lacks boundary markers or sanitization logic to ignore adversarial instructions embedded in the analyzed code. Capability inventory includes shell execution via 'subprocess' and file system access for test generation. No sanitization or validation of the input code is present, creating an exploitable surface for indirect prompt injection.
- [DYNAMIC_EXECUTION]: The assessment logic in 'references/system-assessment.md' dynamically parses and traverses untrusted source code using the 'ast' module, which involves runtime processing of external code structures.
Recommendations
- CRITICAL: 1 file(s) identified as malware by FileRep - DO NOT USE
- AI detected serious security threats
- Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata