segmentfault
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a Python script (
scripts/segmentfault.py) to interact with SegmentFault's API endpoints. The script is designed to run in a restricted environment using only Python's standard library, minimizing the supply chain risk from third-party dependencies.- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted Markdown data from local files or user-provided strings to generate drafts. This creates a surface for indirect prompt injection if the processed content contains instructions that attempt to influence the agent's behavior. - Ingestion points: Markdown content is read via the
--contentor--content-filearguments inscripts/segmentfault.py. - Boundary markers: The skill instructions mandate a dry-run phase and require the
--confirmflag as the final CLI argument to ensure user review before any write operations occur. - Capability inventory: The script has the ability to read arbitrary local files (if paths are provided to the agent) and perform network POST requests to segmentfault.com.
- Sanitization: No explicit filtering or sanitization of the article content is performed before it is sent to the SegmentFault draft gateway.
Audit Metadata