hallmark-build
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run a provided Python script (hallmark_lint.py) and standard project verification tools (lint, build, unit tests) to validate UI changes. This execution is scoped to local development tasks and uses a script included within the skill package.
- [PROMPT_INJECTION]: The skill ingests untrusted repository content such as source code, tokens, and style definitions (SKILL.md). 1. Ingestion points: Repository UI source files and style tokens. 2. Boundary markers: None present. 3. Capability inventory: Execution of local Python linting script and project build tools. 4. Sanitization: Not performed on ingested repository content. While this creates a surface for indirect prompt injection, the risk is assessed as safe due to the absence of high-risk capabilities like network exfiltration or privilege escalation.
Audit Metadata