📋 Context & Background
When building agents designed to handle complex workflows (multi_skill_orchestration), a frequent requirement is pulling fragmented, personal data from isolated workspace environments (e.g., calendar logs, checking account histories, and text exports) and unifying them into a clean, human-readable format.
Currently, agents face optimization friction when executing multiple tool calls sequentially to parse varied file structures (CSV, TXT) without losing context or mixing schema definitions.
🎯 Objective
Implement a robust orchestration pipeline within OpenClaw that allows an agent to seamlessly ingest multiple heterogeneous source files simultaneously, apply explicit filtering/mapping rules, and generate a perfectly aligned, single Markdown table as the final artifact.
📋 Context & Background
When building agents designed to handle complex workflows (multi_skill_orchestration), a frequent requirement is pulling fragmented, personal data from isolated workspace environments (e.g., calendar logs, checking account histories, and text exports) and unifying them into a clean, human-readable format.
Currently, agents face optimization friction when executing multiple tool calls sequentially to parse varied file structures (CSV, TXT) without losing context or mixing schema definitions.
🎯 Objective
Implement a robust orchestration pipeline within OpenClaw that allows an agent to seamlessly ingest multiple heterogeneous source files simultaneously, apply explicit filtering/mapping rules, and generate a perfectly aligned, single Markdown table as the final artifact.