bmad-update
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute several system commands including
git,gh(GitHub CLI),npx, andpython3to perform maintenance, fetch updates, and run validation tests. - [EXTERNAL_DOWNLOADS]: The skill uses
npx --yes bmad-method@6.12.1-next.0to download and execute a specific version of the installer tool from the npm registry. This is the primary mechanism for the skill's update functionality. - [INDIRECT_PROMPT_INJECTION]: The skill derives cleanup actions and module removals by reading project-local configuration files such as
manifest.yamlandskill-manifest.csv. - Ingestion points:
_bmad/_config/manifest.yamland_bmad/_config/skill-manifest.csvare parsed to identify modules and skill directories (SKILL.md). - Boundary markers: No explicit delimiters are used to wrap the data read from these files.
- Capability inventory: The skill performs file system deletions (
rm) and package installations (npx) based on the content of these files. - Sanitization: The instructions include manual sanitization logic by requiring "exact name" matching for deletions to prevent accidental removal of supported components.
Audit Metadata