prototype-hci-pack-beads

Fail

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the bd CLI to create and update issues by interpolating variables like <finding-title>, <description>, and <evidence> directly into shell command strings (e.g., bd create "[MAJOR] <finding-title>"). Since these values are extracted from untrusted repository content, screenshots, and documentation, a malicious file could contain shell metacharacters (e.g., backticks or semicolons) that trigger arbitrary command execution on the host system.\n- [EXTERNAL_DOWNLOADS]: The skill references the beads-cli package and provides a link to the steveyegge/beads repository on GitHub for installation. These are well-known technology resources in the developer community.\n- [PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection through the ingestion of untrusted repository data.\n
  • Ingestion points: Current repository files, screenshots, and documentation processed during the HCI review phase in SKILL.md.\n
  • Boundary markers: Absent; there are no delimiters or specific instructions to the agent to treat external content as untrusted or to ignore embedded commands.\n
  • Capability inventory: The skill performs shell command execution using the bd CLI tool.\n
  • Sanitization: Absent; findings are interpolated into shell commands without validation, escaping, or filtering.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
May 19, 2026, 04:52 PM
Security Audit — agent-trust-hub — prototype-hci-pack-beads