writing-to-logseq

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/write-operations.py includes a main execution block that allows performing Logseq CRUD operations directly via command-line arguments.
  • [PROMPT_INJECTION]: The skill acts as a sink for user-supplied data written to an external application (Logseq). While references/safety-guidelines.md outlines methods for escaping markdown and validating properties, the primary script scripts/write-operations.py does not programmatically enforce these sanitization steps. Ingestion points: create_page, create_block, update_block, append_to_page, and sync_notes methods in scripts/write-operations.py. Boundary markers: None present in the code logic. Capability inventory: Modification of the local Logseq graph via HTTP POST requests to the Editor API. Sanitization: Documented as a best practice but absent in the provided Python implementation.
  • [SAFE]: The skill retrieves configuration data and API tokens from environment variables or a local file at .claude/logseq-expert/env.json, which is a standard and acceptable practice for credential management in this context.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 09:58 AM