confluence

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONNO_CODE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes tasks by calling an external command-line tool named af confluence. This tool is used to perform operations such as searching, reading, and updating Confluence content.
  • [CREDENTIALS_UNSAFE]: The skill requires Atlassian API credentials (JIRA_API_TOKEN, JIRA_EMAIL) to be stored in an environment file. While this is standard for CLI integrations, it involves handling sensitive authentication data.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes data from external Confluence pages and comments which could contain malicious instructions.
  • Ingestion points: Page content and comments are ingested via af confluence get, af confluence search, and af confluence comments.
  • Boundary markers: No specific boundary markers or 'ignore' instructions are provided to separate page content from agent instructions.
  • Capability inventory: The skill can execute CLI commands, read local files via --body-file, and upload attachments.
  • Sanitization: There is no evidence of sanitization or filtering of the content retrieved from Confluence.
  • [NO_CODE]: The skill consists of documentation and instructions for an external CLI tool rather than providing its own executable code.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 03:29 PM