openspec-workflow

Pass

Audited by Gen Agent Trust Hub on Apr 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions guide the user to install the @fission-ai/openspec package via the npm registry to enable enhanced features like validation and dashboards.
  • [COMMAND_EXECUTION]: The agent is authorized to perform file system operations such as creating directories (mkdir) and moving files (mv) within the project's openspec/ directory. It also executes the openspec CLI to perform spec validation and archiving.
  • [PROMPT_INJECTION]: The skill implements a workflow that involves reading and processing specification files (e.g., proposal.md, spec.md). While this represents a surface for indirect prompt injection, the skill mitigates risk by requiring explicit human alignment and approval at every stage before code implementation begins.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 16, 2026, 11:34 AM