ccf-project-scaffolder

Pass

Audited by Gen Agent Trust Hub on Oct 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONOBFUSCATION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill reads existing project artifacts and filesystem structures to guide its initialization process, which introduces a potential surface for processing untrusted instructions.
  • Ingestion points: Existing project files, local directory structures, and ccfa.yaml configuration (SKILL.md, Workflow steps 1 and 4).
  • Boundary markers: The skill relies on artifact-contracts.md to define file ownership and task scope (SKILL.md, Boundaries).
  • Capability inventory: The skill performs file system write operations, including directory creation and template deployment (SKILL.md, Workflow).
  • Sanitization: There are no explicit instructions to validate or sanitize content ingested from existing project files.
  • [OBFUSCATION]: A UTF-8 Byte Order Mark (BOM) was found in a provided template file.
  • Evidence: The file assets/ccfa.yaml begins with the U+FEFF character.
  • Analysis: This character is a common marker used by text editors for identifying file encoding; in this context, it is benign and does not hide malicious payloads or instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 15, 2026, 10:34 PM