legacy-modernizer

Fail

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: CRITICALMETADATA_POISONINGCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [METADATA_POISONING]: Automated reputation scanners identified the documentation URL provided in the skill metadata (https://jeffallan.github.io/claude-skills/skills/specialized/legacy-modernizer/) as blacklisted.
  • [COMMAND_EXECUTION]: Automated file reputation scanners flagged the skill as malware (FileRepMalware) due to code in references/system-assessment.md that utilizes the subprocess module to execute git commands on the host system.
  • [INDIRECT_PROMPT_INJECTION]: The skill architecture for legacy system analysis creates a significant surface for indirect prompt injection attacks.
  • Ingestion points: The LegacyCodeAnalyzer in references/system-assessment.md programmatically reads the content of all Python files within a user-specified project directory.
  • Boundary markers: The instructions lack delimiters or safety guardrails to prevent the agent from executing instructions maliciously embedded within analyzed codebases.
  • Capability inventory: The skill tools include subprocess execution (references/system-assessment.md), network proxying via httpx (references/migration-strategies.md), and file system writes for snapshot storage (references/legacy-testing.md).
  • Sanitization: The skill does not implement validation or filtering logic for ingested source code before it is parsed and processed by the agent.
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
Risk Level
CRITICAL
Analyzed
Sep 14, 2026, 09:34 PM
Security Audit — agent-trust-hub — legacy-modernizer