The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Jira tickets, including titles, descriptions, acceptance criteria, and user comments, which are materialized into local Markdown files (e.g., story.md, comments.md) via the bun run jira:sync-issues tool.
Ingestion points: Data is fetched from Jira and stored in files within the <PBI_FOLDER> (references/session-entry-points.md).
Boundary markers: There are no explicit boundary markers or instructions provided to the agent or sub-agents to ignore potential commands or prompt injections embedded within the ticket content.
Capability inventory: The skill utilizes the /acli and /xray-cli tools for Jira writes, executes authenticated network requests via curl, performs database queries via the DBHub MCP, and automates browser interactions using Playwright (SKILL.md, references/exploration-patterns.md).
Sanitization: There is no evidence of sanitization or filtering of the ingested external content before it is processed by the AI to generate test plans (ATP), test cases, and reports.

sprint-testing