openspec-new-change

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes subcommands of the openspec CLI, such as new change, status, and instructions, to scaffold projects and retrieve metadata based on user input.
  • [PROMPT_INJECTION]: The skill ingests untrusted user input via the AskUserQuestion tool, creating a surface for indirect prompt injection attacks.
  • Ingestion points: User descriptions captured in SKILL.md.
  • Boundary markers: Absent; the skill lacks explicit delimiters for separating user input from instructions.
  • Capability inventory: Shell command execution capabilities for the openspec CLI.
  • Sanitization: The skill provides a guardrail instructing the agent to derive a kebab-case name and validate it, prompting the user for correction if the format is invalid.
  • [EXTERNAL_DOWNLOADS]: The skill documentation specifies that the openspec CLI is a prerequisite. No logic is included to automatically download or install external code or dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 11:12 AM
Security Audit — agent-trust-hub — openspec-new-change