The Agent Skills Directory

[COMMAND_EXECUTION]: Utilizes curl with subshell-based Base64 encoding to handle authentication headers for Atlassian's REST APIs.
[EXTERNAL_DOWNLOADS]: Fetches ticket and documentation data from api.atlassian.com. This is a well-known service, and the operations are consistent with the skill's primary purpose.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection attack surface because it retrieves and processes external content.
Ingestion points: The skill reads data from Jira issue descriptions, comments, and Confluence page bodies into the agent's context (defined in SKILL.md).
Boundary markers: External content is interpolated into the context without specific delimiters or "ignore previous instructions" warnings.
Capability inventory: All operations are limited to network requests via curl to Atlassian's API gateway; there are no file-writing or arbitrary code execution capabilities.
Sanitization: While HTML content is converted to markdown for formatting, there is no validation or filtering implemented to detect or strip instructions embedded within the retrieved data.

atlassian-reader