skills-index-snippets
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions to execute local shell scripts for repository maintenance, specifically
./scripts/generate-skill-index-snippets.shand./scripts/validate-marketplace.sh. These actions are consistent with the skill's stated purpose of managing skill indexes. - [INDIRECT_PROMPT_INJECTION]: The skill describes a process for ingesting metadata from repository files to generate documentation snippets, which represents a potential attack surface for indirect prompt injection if those files contain malicious instructions.
- Ingestion points: Metadata from
.claude-plugin/plugin.json, frontmatter fromSKILL.mdfiles, and markdown files within theagents/directory. - Boundary markers: The skill does not explicitly define or enforce boundary markers or delimiters for the ingested metadata in its generation instructions.
- Capability inventory: The skill facilitates the execution of local shell scripts for metadata validation and generation.
- Sanitization: There are no documented sanitization or validation routines for the metadata fields (e.g.,
nameordescription) before they are incorporated into the generated snippets.
Audit Metadata