The Agent Skills Directory

[COMMAND_EXECUTION]: The adr-drift.sh script executes grep -rE using regex patterns sourced from ADR markdown files to identify discrepancies between documentation and code.\n- [DATA_EXFILTRATION]: The skill transmits aggregated project data to well-known external services, including Slack, Notion, and Google APIs, using environment-provided API tokens. This behavior is documented and scoped to the intended function of the skill.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data that could contain malicious instructions.\n
Ingestion points: Data is pulled from Linear tickets, GitHub PR descriptions, Granola notes, and Google Drive documents via various gather-*.sh scripts.\n
Boundary markers: The skill does not employ explicit delimiters or system instructions to ignore embedded commands within the ingested data during synthesis.\n
Capability inventory: The skill has network access for distribution and filesystem write permissions for local storage.\n
Sanitization: A sanitize.sh script provides mitigation by redacting customer names and specific URLs before the briefing is sent to Slack or Notion.

morning-briefing