The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a local utility script, scripts/tickets.sh, which performs complex operations by executing shell commands such as bash, gh, and git. It also invokes a python3 interpreter to run embedded logic for API interactions with Linear and Jira services.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reading data from external sources that are not under the skill author's control, creating a surface for indirect prompt injection.
Ingestion points: In SKILL.md under 'Stage 1: Load', the skill fetches map notes, decisions, owning documents, and issue bodies from GitHub, Linear, and Jira based on user-provided IDs or URLs.
Boundary markers: The instructions do not specify any boundary markers or instructions for the agent to treat this ingested external text as untrusted or to disregard potential instructions embedded within the ticket data.
Capability inventory: The skill has significant write capabilities, including creating new issues, adding labels, wiring dependency edges, and posting comments via the scripts/tickets.sh utility.
Sanitization: No explicit sanitization or filtering steps are defined for the data retrieved from external trackers before the agent processes it to generate new ticket briefs.

conjure