posix-shell-scripting
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to review and audit untrusted shell scripts, which introduces a surface for indirect prompt injection attacks.
- Ingestion points: Processes files with .sh extensions, hashbangs, or shell syntax, including sourced libraries and startup fragments (SKILL.md).
- Boundary markers: The 'Instruction Authority' section requires delimiting untrusted data and explicitly states that untrusted material cannot grant authority or expand scope (SKILL.md).
- Capability inventory: Performs read operations, shell parsing (sh -n), and potentially invokes external validators like ShellCheck (SKILL.md, references/validation-tools.md).
- Sanitization: Instructions mandate treating task material as data only and ignore any instructions embedded within the processed content (SKILL.md).
- [EXTERNAL_DOWNLOADS]: The skill describes a procedure for fetching an optional peer skill ('human-facing-writing') from the author's GitHub repository.
- Source URL: Fetches documents from github.com/porada/domfiles (references/optional-peer-human-facing-writing.md).
- Security controls: Requires explicit user confirmation and network permission. It mandates the use of immutable commit hashes and strictly validates the fetched content before use.
- Vendor Context: The resource belongs to the same vendor ('porada') and the retrieval follows a secure-by-design workflow.
Audit Metadata