sdd-init

Pass

Audited by Gen Agent Trust Hub on Mar 27, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill represents an attack surface for indirect prompt injection as it reads and processes multiple untrusted project-level files to build its configuration and skill registry.
  • Ingestion points: Reads package.json, go.mod, pyproject.toml, agents.md, CLAUDE.md, .cursorrules, and other project metadata files in SKILL.md.
  • Boundary markers: None identified; detected context is directly interpolated into openspec/config.yaml and .atl/skill-registry.md.
  • Capability inventory: Performs file system writes (directory/file creation) and utilizes the mem_save tool for context persistence as seen in SKILL.md.
  • Sanitization: No explicit sanitization or validation of the content read from the project files is described.
  • [DATA_EXFILTRATION]: The skill scans user-level directories (e.g., ~/.claude/skills/, ~/.config/opencode/skills/) to build a skill registry. While this involves reading from the home directory, the behavior is constrained to specific tool-related subdirectories and used only for local configuration purposes, not external transmission.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 27, 2026, 10:57 AM