tree-sitter-language-pack

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download of parser libraries for 306 programming languages. These downloads are performed on-demand or via the ts-pack download command, fetching resources from the author's official GitHub repository (xberg-io/tree-sitter-language-pack).
  • [REMOTE_CODE_EXECUTION]: The installation instructions include patterns for executing remote code from package registries, such as npx @xberg-io/ts-pack-cli and uvx --from ts-pack-cli ts-pack. These are standard methods for utilizing the vendor's CLI tools.
  • [COMMAND_EXECUTION]: The skill provides numerous shell commands for parsing, processing, and managing code intelligence, which are necessary for its documented functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external source code files through the parse and process commands, representing an ingestion surface for potentially untrusted data.
  • Ingestion points: Files passed to ts-pack parse <file> and ts-pack process <file> in SKILL.md.
  • Boundary markers: None identified in the CLI output format.
  • Capability inventory: The tool can read local files, execute shell commands, and download grammar files from the vendor's repository.
  • Sanitization: Not explicitly defined for the parsed output, which is typical for a low-level parsing library.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 10:22 PM
Security Audit — agent-trust-hub — tree-sitter-language-pack