skills/tenzir/ship/tenzir-ship/Gen Agent Trust Hub

tenzir-ship

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes several command-line tools including git, gh (GitHub CLI), and a local bash script detect-change-scope.sh. These are used for standard repository operations like checking for changes, managing tags, and triggering GitHub Actions workflows.
  • [EXTERNAL_DOWNLOADS]: The skill uses uvx to run the tenzir-ship package. This is a vendor-provided tool (Tenzir) used for the skill's primary purpose of release management. The downloads are from standard registries and target official vendor software.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from git diff outputs and commit messages to generate changelog summaries.
  • Ingestion points: Git diff output and commit history processed by detect-change-scope.sh and instructions in add-changelog-entry.md.
  • Boundary markers: Not explicitly defined in the prompts when summarizing diffs.
  • Capability inventory: The skill has the ability to write to the file system (changelog entries) and execute shell commands via uvx and gh.
  • Sanitization: The skill uses temporary files and the --description-file flag to handle multi-line content safely, reducing the risk of shell injection, though it remains theoretically susceptible to instruction injection within the processed text.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 01:51 PM
Security Audit — agent-trust-hub — tenzir-ship