sdlc-import
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local system commands including
git,grep, and a local Python tool (.wiki-llm/tools/wiki_llm.py) to manage the wiki repository. These operations are scoped to the project directory and are standard for the skill's intended purpose. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data sources such as source code folders and converted documents. This risk is mitigated by a multi-stage human approval process (Intent, Design, and Change-set gates) where every generated file must be reviewed and explicitly approved by the user before being committed.
- [DATA_EXPOSURE]: The skill includes proactive security instructions to prevent sensitive data exposure. It explicitly directs the agent to never copy secrets, credentials, or personal data from the import source into the wiki, recommending the use of setting names instead of values.
Audit Metadata