openstack-kolla-ansible-ops
Pass
Audited by Gen Agent Trust Hub on Jun 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on powerful administrative tools to manage infrastructure, including kolla-ansible, docker, and the openstack CLI. These tools allow for deep system changes and container manipulation.
- Evidence: Commands like 'kolla-ansible reconfigure', 'docker exec', and 'openstack compute service set' are core to the operational workflows described in SKILL.md.
- [EXTERNAL_DOWNLOADS]: The skill provides procedures for software lifecycle management that involve fetching packages from official registries.
- Evidence: The major upgrade procedure instructs the user to run 'pip install --upgrade kolla-ansible', which downloads software from the standard Python Package Index (PyPI).
- [DATA_EXFILTRATION]: The skill interacts with sensitive file paths containing credentials and cryptographic materials. While no network exfiltration was detected, the access to these files is a security-sensitive operation.
- Evidence: Instructions involve editing '/etc/kolla/passwords.yml' for rotation and placing certificates in '/etc/kolla/certificates/'.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes external configuration files (YAML and INI) and command outputs without explicit sanitization or boundary markers.
- Ingestion points: Reads from '/etc/kolla/globals.yml', '/etc/kolla/passwords.yml', and various configuration overrides in '/etc/kolla/config/'.
- Boundary markers: Not present.
- Capability inventory: Includes full infrastructure reconfiguration, container restarts, and executing commands inside containers via 'docker exec'.
- Sanitization: No validation or escaping of the configuration content before processing is documented.
Audit Metadata