atlassian-cli
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides detailed instructions for executing the Atlassian CLI (
acli) binary to perform various operations, including Jira work item management, project administration, and organization-level user lifecycle tasks. - [INDIRECT_PROMPT_INJECTION]: The skill documents commands that ingest and process external data, creating an attack surface for indirect prompt injection via untrusted inputs.
- Ingestion points: User-supplied JQL strings (
--jql), saved filter IDs (--filter), and local files (--from-json,--from-file) as described inSKILL.mdandreferences/workitem.md. - Boundary markers: There are no instructions for implementing delimiters or security warnings when interpolating these data sources into agent commands.
- Capability inventory: The skill facilitates powerful capabilities including account deactivation, project deletion, and bulk issue updates.
- Sanitization: No sanitization or validation mechanisms are described for the external data ingested by the CLI tools.
- [SAFE]: The documentation includes security best practices for authentication, specifically recommending that API tokens be passed via standard input to prevent them from being logged in persistent shell history files.
Audit Metadata