easysam-skill
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill incorporates a 'Deployment Failure Circuit-Breaker' rule in
SKILL.mdandreferences/troubleshooting.md, instructing the agent to stop automated retries and alert the user after two failures to prevent resource exhaustion or cost spikes. - [REMOTE_CODE_EXECUTION]: The
assets/publish.ymlfile contains a command to fetch and execute theuvinstallation script from the officialastral.shdomain. This is a common setup procedure for this specific well-known developer tool. - [COMMAND_EXECUTION]: The agent is instructed to use the
uvrunner to execute various CLI operations for theeasysamtool, includinginit,generate,deploy, andinspectcommands for project management. - [EXTERNAL_DOWNLOADS]: The skill's CI/CD pipeline template (
assets/publish.yml) utilizes official GitHub Actions from AWS (aws-actions/configure-aws-credentials) and retrieves runtime dependencies from standard Python registries. - [PROMPT_INJECTION]: The skill includes instructions in the YAML frontmatter to prioritize its own workflows for any serverless scaffolding requests, ensuring the agent adheres to the 'EasySAM' modular patterns even when not explicitly requested.
Audit Metadata