developer-growth-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill accesses sensitive information by reading
~/.claude/history.jsonl, which contains private user prompts, project details, and potentially pasted credentials or source code. This data is processed and subsequently transmitted to an external Slack workspace using the Rube MCP toolset. While the instructions state the report is sent to the user's own DMs, this establishes a data flow from local sensitive files to a remote network endpoint. - [INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted data from the chat history file which could contain attacker-controlled instructions.
- Ingestion points: Chat history is read from
~/.claude/history.jsonl(SKILL.md). - Boundary markers: No delimiters or "ignore embedded instructions" warnings are defined to separate the log data from the agent's analysis instructions.
- Capability inventory: The skill possesses the capability to read local files, perform external searches via HackerNews, and send messages to Slack (SKILL.md).
- Sanitization: The instructions do not include steps to sanitize, filter, or escape the content retrieved from the chat history before it is analyzed or included in the Slack report.
Audit Metadata