write-agentsmd

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (wc, grep, awk) and a small Python script to analyze local files.
  • Evidence: Shell commands in the 'Bloat Scan' section of SKILL.md execute logic on the local file system.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it analyzes external data (the contents of AGENTS.md or CLAUDE.md) that may contain malicious instructions.
  • Ingestion points: The skill reads and classifies content from AGENTS.md.
  • Boundary markers: Absent. The skill does not instruct the agent to treat the file content as untrusted data or use delimiters.
  • Capability inventory: The skill utilizes subprocess execution for analysis and has the capability to write files.
  • Sanitization: Absent. The skill lacks logic to filter or escape instructions found within the analyzed files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 04:34 PM
Security Audit — agent-trust-hub — write-agentsmd