atlassianapi
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- PROMPT_INJECTION (LOW): Indirect Prompt Injection surface detected (Category 8). The skill processes untrusted data from external sources and possesses the capability to modify Jira issues and Confluence pages.
- Ingestion points: The
sync_pr_to_jirafunction ingests GitHub PR data, andhandle_jira_webhookprocesses incoming webhook JSON payloads. - Boundary markers: Absent; external strings such as PR titles and descriptions are directly interpolated into API calls for comments and page creation.
- Capability inventory: High-impact write capabilities including
jira.add_comment,jira.transition_issue, andconfluence.create_pageinSKILL.md. - Sanitization: Absent; the code snippets do not demonstrate escaping or validation of external content before it is used in automation recipes.
Audit Metadata