Deep Research

This skill is internally consistent with its stated purpose: it orchestrates parallel sub-agents, collects multi-source findings, synthesizes analysis, and produces Markdown/PDF reports. Before installing or running it, consider the following: - Review and confirm the workspace path you want the skill to use (SKILL.md uses research-orchestrator/... while the Python code uses deep-research/...); files will be created under that workspace and the mismatch can cause confusion. - The PDF converter may run 'npm install -g md-to-pdf' and uses 'npx md-to-pdf'. If you prefer not to allow global npm installs, ensure the environment already has md-to-pdf or run the conversion in a controlled environment. - The skill will spawn parallel sub-agents (sessions_spawn) and perform many searches; expect increased network activity and potentially many created files. If you have quotas, rate limits, or monitoring needs, sandbox or monitor the agent's activity. - No credentials are requested by the skill, which is good. Still review the code if you will run it in an environment that contains sensitive files — the scripts write to the workspace and may read/write JSON/MD files there. If you want extra caution: run the skill in an isolated testing workspace, verify that required binaries (python3, node/npm, curl) are present and acceptable to use, and inspect the generated task directories before granting broader access.