service-blueprint
Warn
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill performs a download of an external repository from GitHub (
https://github.com/j-clegg/service-blueprint-skill.git) that does not belong to a recognized trusted organization. - [COMMAND_EXECUTION]: The skill executes shell commands to perform its manual installation and verification, including
git clone,grep, and file system tests ([ -e ... ]). - [REMOTE_CODE_EXECUTION]: The skill instructs the agent to read and 'follow' the instructions located in the externally cloned
SKILL.md. For an AI agent, following natural language instructions from an unverified remote source is equivalent to executing unvetted code. - [INDIRECT_PROMPT_INJECTION]: The skill defines a large surface for indirect prompt injection by ingesting and following content from multiple external files (
METHODOLOGY.md,SCHEMA.md, andRENDERING.md). - Ingestion points: Files cloned from
j-clegg/service-blueprint-skillrepository. - Boundary markers: None identified; the agent is instructed to simply follow the files.
- Capability inventory: Shell command execution, file system access, and repository cloning.
- Sanitization: No validation or sanitization is performed on the content of the external files before the agent processes them.
Audit Metadata