synthesis-slack-sync

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted content from Slack messages, which can be used for indirect prompt injection.\n
  • Ingestion points: Messages are read from Slack via MCP tools (slack_read_channel, slack_read_thread) and written to local transcript files and daily action plans.\n
  • Boundary markers: The skill uses Markdown headers and dividers for structure but lacks explicit security delimiters or 'ignore' instructions for the ingested message content.\n
  • Capability inventory: The agent has file system write access to specific repository paths and can execute Slack MCP tools.\n
  • Sanitization: Content is transcribed verbatim into transcripts without validation or filtering of potential control sequences or instructions.\n- [EXTERNAL_DOWNLOADS]: The provided Python script retrofit_permalinks.py requires the PyYAML package for parsing configuration files, which is a standard library dependency.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 09:44 AM
Security Audit — agent-trust-hub — synthesis-slack-sync