breadcrumbs-development-workflow

Pass

Audited by Gen Agent Trust Hub on Apr 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes multiple shell commands including git, grep, find, and ripgrep (rg) to navigate the vault structure, verify terminology, and manage the commit workflow.
  • [COMMAND_EXECUTION]: The skill invokes a project-specific command caliber refresh to synchronize agent instructions and documentation metadata before committing changes.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it is designed to read, edit, and search content within a documentation vault that directly influences agent behavior through files like CLAUDE.md, AGENTS.md, and .cursorrules.
  • Ingestion points: Markdown documentation files located throughout the breadcrumbs-docs-vault/ structure (e.g., Concepts.md, Guides/).
  • Boundary markers: No specific boundary markers or "ignore instructions" delimiters are used when processing or searching vault content.
  • Capability inventory: The skill can execute shell commands (git, rg, find, caliber) and perform file system modifications.
  • Sanitization: There is no evidence of sanitization or validation of the Markdown content before it is committed or used to update agent configurations.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 16, 2026, 02:49 PM