The Agent Skills Directory

[SAFE]: The skill contains no executable code, scripts, or binary files, consisting entirely of markdown instructions and metadata schemas.
[NO_CODE]: No code was detected in the skill file.
[PROMPT_INJECTION]: An indirect prompt injection surface exists as the skill processes external Notion data for local file creation.
Ingestion points: External data from Notion databases like Projects, Roadmap, and Engineering Specs.
Boundary markers: No delimiters or instructions to ignore embedded content are specified.
Capability inventory: File writing and folder creation within the pm-workspace-docs/ directory.
Sanitization: No explicit data validation or sanitization steps are defined for external strings.
[DATA_EXFILTRATION]: No unauthorized data transmission or credential exposure was found. Notion Database IDs are included for mapping but are not sensitive secrets.
[EXTERNAL_DOWNLOADS]: The skill mentions Notion and Composio MCP tools, which are well-known and legitimate technology services.

notion-sync