Skills
skills/lucastamoios/celeiro/openspec-to-beads/Gen Agent Trust Hub

openspec-to-beads

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (LOW): The skill uses templates in templates/issue-creation.md to construct and execute shell commands (e.g., bd create). It interpolates task descriptions and categories directly from local files (tasks.md) without escaping characters. A malicious project file could use shell metacharacters to execute unauthorized commands.
  • [PROMPT_INJECTION] (LOW): Vulnerable to indirect prompt injection (Category 8). The skill is instructed to read and 'understand' untrusted content from openspec/changes/ to drive its logic. 1. Ingestion points: tasks.md, proposal.md, spec.md, and design.md. 2. Boundary markers: Absent. 3. Capability inventory: Full access to bd (issue tracking) and openspec CLI tools. 4. Sanitization: Absent; the agent is directed to bridge gaps 'intelligently' without specific safety filters for ingested content.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:58 PM