speckit-opsmill-prep

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses a web fetching tool in Phase 5b to retrieve content from external URLs provided in the user's arguments (such as Notion, GitHub issues, or documentation sites) to serve as a source of truth for requirement alignment.
  • [COMMAND_EXECUTION]: The skill coordinates a complex workflow by invoking multiple specialized sub-skills, including speckit-specify, speckit-plan, speckit-critique-run, speckit-tasks, and speckit-checkpoint-commit, to manage the project lifecycle.
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface where external, untrusted content is processed autonomously.
  • Ingestion points: Content is ingested from user-provided URLs and text in the $ARGUMENTS variable in Phase 5.
  • Boundary markers: No explicit security delimiters or "ignore instructions" warnings are applied to the fetched content before processing.
  • Capability inventory: The skill has the capability to trigger autonomous updates to local project files (spec.md, plan.md, tasks.md) by re-invoking the specification and planning sub-skills.
  • Sanitization: The skill lacks sanitization or validation of the fetched PRD content, potentially allowing embedded malicious instructions to influence the generated project plan or task list.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 09:09 PM
Security Audit — agent-trust-hub — speckit-opsmill-prep