skills/ycmjason/tixmd/tixmd-groom/Gen Agent Trust Hub

tixmd-groom

Warn

Audited by Gen Agent Trust Hub on Mar 22, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes the command 'pnpx @tixmd/cli list' to retrieve ticket data, which involves executing code from an external package.
  • [EXTERNAL_DOWNLOADS]: The 'pnpx' command automatically downloads the '@tixmd/cli' package from the npm registry at runtime. As the package is unversioned and not from a verified vendor, it poses a supply chain risk.
  • [PROMPT_INJECTION]: The skill reads content from '.tixmd/tixs/*.md' and '.tixmd/project.md' to guide its refinement process, creating an indirect prompt injection surface.
  • Ingestion points: Ticket markdown files and project configuration files.
  • Boundary markers: None identified; the agent is instructed to process the body and frontmatter directly.
  • Capability inventory: Shell command execution via pnpx and filesystem read/write access.
  • Sanitization: No explicit sanitization or validation of the ingested file content is performed.
  • [DATA_EXFILTRATION]: The skill constructs file paths using user-provided arguments (e.g., '.tixmd/tixs/.md') without input validation. This could allow an attacker to use path traversal sequences (like '../../') to trick the agent into reading or writing sensitive files outside the intended directory.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 22, 2026, 05:34 PM
Security Audit — agent-trust-hub — tixmd-groom