The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration were detected. The skill's behavior aligns with its stated purpose of drafting status updates.
[COMMAND_EXECUTION]: The skill uses shell commands (date, linear, jq) to perform legitimate tasks such as calculating date windows and querying issue data. These tools are explicitly listed in the allowed-tools configuration.
[PROMPT_INJECTION]: The skill processes data from an external source (Linear issue titles), which constitutes an indirect prompt injection surface.
Ingestion points: Issue titles and identifiers retrieved via the linear CLI in SKILL.md.
Boundary markers: The draft is explicitly isolated within markdown code blocks for user inspection.
Capability inventory: Subprocess execution is restricted to the specific tools required for data retrieval and formatting.
Sanitization: The skill includes logic to replace characters (|, >) that would otherwise interfere with Slack's link syntax or the drafting process.

daily-update