atlas-setup
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches configuration manifests from the vendor's official domain (
atlas.logbookfordevs.com) and references agent skill packages from GitHub repositories belonging to the vendor and established community developers.- [COMMAND_EXECUTION]: The skill utilizes command-line tools includingnpx skillsand@logbookfordevs/afkto perform host inventory checks and manage the installation state of the environment.- [REMOTE_CODE_EXECUTION]: The skill automates the installation of agent skills from remote GitHub repositories. This behavior is the primary intended function of the skill and is mitigated by explicit instructions requiring the agent to obtain user authorization before executing any installation or removal commands.- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from JSON manifests and dependency registries. While this creates an ingestion surface for untrusted data, the skill instructions emphasize verifying skill identities and source alignment, and the operations are restricted to structured configuration fields.- [DYNAMIC_EXECUTION]: Shell commands for managing skills are constructed at runtime based on the contents of the local and remote manifests. The skill minimizes risk by requiring the user to review the concrete targets and scope of these commands before they are run.
Audit Metadata