skills/maroffo/claude-forge/bujo-sync/Gen Agent Trust Hub

bujo-sync

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses local configuration files (.mcp.json and Journal/.bujo-sync-state.json) for legitimate task management and synchronization purposes.\n- [SAFE]: External communication is restricted to configured MCP servers (ClickUp and Linear), following standard protocols for agent-tool interaction.\n- [SAFE]: The skill incorporates safety boundaries by using explicit HTML comment markers to isolate external task data within user documents, preventing the agent from modifying content outside the designated sync zones.\n- [SAFE]: An indirect prompt injection surface is present as the skill processes external task metadata from ClickUp and Linear.\n
  • Ingestion points: Task titles and statuses are fetched via MCP tools as described in SKILL.md.\n
  • Boundary markers: The skill utilizes <!-- bujo-sync:...:start --> and <!-- bujo-sync:...:end --> markers to delimit untrusted content.\n
  • Capability inventory: Capabilities include file system read/write access for journal management and MCP tool execution for status updates (SKILL.md, references/sync-details.md).\n
  • Sanitization: While explicit sanitization is not mentioned, the skill mandates strict formatting for synced tasks, which helps constrain the interpretation of external data.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 07:15 AM