developer-growth-analysis
Warn
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: MEDIUMDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: Accesses the local file
~/.claude/history.jsonl. This file stores a complete history of user interactions, including any code snippets, project architectures, or sensitive information pasted into the chat interface. - [DATA_EXFILTRATION]: The skill transmits the analyzed contents to Slack using the
RUBE_MULTI_EXECUTE_TOOL. This creates a data flow where private local history is processed and sent to a third-party messaging service. - [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection.
- Ingestion points: Reads and processes data from
~/.claude/history.jsonl, which contains untrusted input from previous chat sessions. - Boundary markers: Absent. The skill does not use delimiters to isolate the history content from its own instructions.
- Capability inventory: Includes the ability to search external resources (HackerNews) and post messages to Slack.
- Sanitization: Absent. There is no evidence of filtering or sanitizing the chat history to prevent the agent from obeying instructions that might be embedded in the logs.
Audit Metadata