The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script (thread_checker.py) to analyze transcript files and identify threads that require re-syncing. This script uses standard Python library modules and operates solely on local files provided via command-line arguments.
[DATA_EXPOSURE]: The skill reads Slack messages and writes them to local markdown transcript files in the user's home directory. It also manages a local configuration file (.claude/slack-sync.yaml) and an action plan. This file access is intrinsic to the skill's purpose and does not involve exfiltration to external domains.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from Slack, which is an external source that could contain malicious instructions.
Ingestion points: Slack message content ingested via slack_read_channel and slack_read_thread MCP tools.
Boundary markers: Transcript files use structured markdown headers (## #channel-name, ### [Author]) to delimit different messages and metadata.
Capability inventory: The agent can read/write local files and execute the provided thread_checker.py script.
Sanitization: The protocol requires the agent to use a strict 'Draft Message Format' and verify facts against primary sources (e.g., source code, deploy logs) before drafting replies, which serves as a verification layer against potentially malicious data in Slack threads.

synthesis-slack-sync