The Agent Skills Directory

[DATA_EXFILTRATION]: The skill reads sensitive local data from ~/.claude/history.jsonl, which contains full transcripts of previous interactions, project details, and code snippets from pastedContents. This information is then processed and sent to an external service (Slack) via the RUBE_MULTI_EXECUTE_TOOL. While this is the primary purpose of the skill, it involves the transmission of private development history to a third-party platform.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from local chat logs which may contain malicious instructions designed to influence the agent's behavior during analysis.
Ingestion points: The agent reads the display (user messages) and pastedContents fields from ~/.claude/history.jsonl.
Boundary markers: None are present. The instructions do not define delimiters or provide specific 'ignore' directives for the content found in the history file.
Capability inventory: The skill has the ability to send messages to Slack (RUBE_MULTI_EXECUTE_TOOL) and perform web searches on HackerNews (RUBE_SEARCH_TOOLS).
Sanitization: No sanitization or filtering is performed on the ingested chat history before it is analyzed for patterns and problem types.
[EXTERNAL_DOWNLOADS]: The skill uses RUBE_SEARCH_TOOLS to query and retrieve content from HackerNews. While searching a well-known service is generally acceptable, it introduces an external dependency on third-party article content.

developer-growth-analysis