write-korean-blog
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run local Python scripts using the
uv runcommand to perform linting, outlining, and article checks on generated content. - [INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied topics, data, and drafts, creating a potential attack surface where malicious input could influence agent behavior. • Ingestion points: Data and topics provided by the user for blog generation or reconstruction as described in the skill metadata. • Boundary markers: None identified; user content is processed without explicit delimiters or instructions to isolate it from the skill's logic. • Capability inventory: The skill has the ability to execute shell commands and interact with the file system for writing and checking articles. • Sanitization: There is no evidence of input validation or sanitization before processing user-provided content.
- [DYNAMIC_EXECUTION]: The skill uses environment variables to construct script paths for execution (e.g.,
$CORE_SKILL_DIR/scripts/lint.py), which involves dynamic path resolution at runtime.
Audit Metadata