requirements-verification

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands including make check-openspec, make setup, and the openspec validation binary located in node_modules. These commands are used to validate the structural integrity of specification files and setup the environment. While specific to the OpenSpec workflow, this capability allows for the execution of arbitrary commands if inputs are manipulated.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data which could lead to unintended behavior if the processed files contain malicious instructions.
  • Ingestion points: Markdown specification files located at openspec/specs/<capability>/spec.md and delta specs under openspec/changes/**/specs/.
  • Boundary markers: The skill uses markdown headers such as ### Requirement: and #### Scenario: to delimit content, but it does not specify sanitization or explicit instructions to ignore embedded natural language commands.
  • Capability inventory: The agent can read files from the internal/ source code directory and execute shell commands using the make utility and the openspec CLI.
  • Sanitization: No explicit sanitization or filtering of the content within the markdown requirement documents is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 02:55 AM