zhshie

Downstream — Development skill for Claude Code

Development community

Turn a SUCCEEDED run's outputs into figures or further analysis.

How to install Downstream

Installs to ~/.claude/skills/zhshie-agentic-bioflow_v2-downstream/SKILL.md

Terminal
mkdir -p ~/.claude/skills/zhshie-agentic-bioflow_v2-downstream && curl -fsSL https://raw.githubusercontent.com/zhshie/agentic-bioflow_v2/HEAD/commands/downstream.md -o ~/.claude/skills/zhshie-agentic-bioflow_v2-downstream/SKILL.md

Restart Claude Code, or start a new session, for it to be picked up.

What Downstream does


description: Turn a SUCCEEDED run's outputs into figures or further analysis argument-hint: [run id or path to results]

Paths below such as `scripts/...` and `docs/...` are this plugin's own files, never the user's working directory. Installed as a plugin they are under `${CLAUDE_PLUGIN_ROOT}`; read them straight from the repository otherwise.

This picks up where `runs` leaves off — after a run SUCCEEDED and its outputs have been delivered, not before. It carries **no pipeline-specific

Alternatives in Development

  • Claude Context Mode — An MCP server that sits between Claude Code and these outputs 5.3k ★
  • Setup Context — Set up your Skill Memory — a pm-context.md every skill reads so outputs come back tailored to you 1.3k ★
  • Sub Skill Figures — Sub-skill subagent prompt — 调女娲.skill 蒸馏 top figure 126 ★

Full documentation available on GitHub

View Source Repository