algo-reconstruct

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes a local script (scripts/compile_algo.py) that uses subprocess.run() to execute LaTeX engines (xelatex or pdflatex) to compile generated code into PDF files. This constitutes runtime compilation and execution of generated content.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted user-provided content without sufficient safeguards.
  • Ingestion points: The skill reads external paper methodology sections, appendices, and source code files as specified in the SKILL.md workflow.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat ingested data as untrusted or to ignore instructions within it.
  • Capability inventory: The agent has permissions to read files, write .tex files, and execute a shell script for compilation.
  • Sanitization: No input validation or filtering is performed on the content gathered from papers or code repositories before it is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 06:24 PM
Security Audit — agent-trust-hub — algo-reconstruct