The Agent Skills Directory

Handling of Untrusted External Content: The skill processes information from external sources that are not under the direct control of the user, specifically Gmail threads, Slack messages, and customer support conversations.
Ingestion points: Data is ingested through tool calls like search_threads for Gmail and Slack search, as specified in reference/data_sources.md.
Boundary markers: The instructions do not define specific delimiters or warnings to ignore potentially conflicting instructions embedded within these external messages.
Capability inventory: The skill possesses the ability to write to the file system and post summaries to Slack (as described in Step 5 of SKILL.md).
Sanitization: No specific sanitization or filtering logic for the content of messages is detailed in the synthesis instructions.
Automated Data Access: The skill is configured to automatically pull data from connected financial and communication tools upon invocation without requiring per-run user confirmation.
This design is intended to minimize latency and improve the user experience for a reporting tool, granting the agent immediate access to the configured data sources once triggered (documented in SKILL.md and reference/gotchas.md).
A safety checkpoint is included for writing operations, requiring explicit user approval before posting data to external communication channels such as Slack.

business-pulse