bmad-shard-doc

Warn

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires downloading the '@kayvan/markdown-tree-parser' package from the NPM registry at runtime via 'npx'.
  • [REMOTE_CODE_EXECUTION]: The execution of an unverified third-party package directly from a public registry poses a risk as the downloaded code is executed on the local environment without prior verification.
  • [COMMAND_EXECUTION]: The skill constructs shell commands by interpolating user-provided inputs for '[source-document]' and '[destination-folder]'. This pattern is vulnerable to command injection if the provided paths contain shell metacharacters (e.g., semicolons or backticks).
  • [COMMAND_EXECUTION]: The workflow includes functionality to delete the original source document from the file system, which is a destructive operation.
  • [PROMPT_INJECTION]: The skill processes untrusted external markdown content which could contain malicious instructions designed to hijack the agent's logic during the parsing phase.
  • Ingestion points: The content of the document provided at '[source-document]' in workflow.md.
  • Boundary markers: No delimiters or isolation techniques are used to separate the external file content from the agent's operational instructions.
  • Capability inventory: Shell command execution via 'npx' and file system operations (move/delete) defined in workflow.md.
  • Sanitization: There is no evidence of sanitization or safety filtering performed on the markdown content before it is processed by the external tool.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 16, 2026, 06:48 AM