# Eda Autoidea Bridge

> Bridge autoIdea literature recommendation, historical formulation, and idea-generation outputs into the current repo's paper/knowledge/skill workflow.

- Skill: `mr-fang-vlsi/eda-autoidea-bridge` (Agent Skill, multi-file: 7 files)
- Install (CLI): `npx skillmds@latest add mr-fang-vlsi/eda-autoidea-bridge`
- Raw SKILL.md: https://api.skillmd.com/api/skills/mr-fang-vlsi/eda-autoidea-bridge/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: Mr-Fang-VLSI (https://skillmd.com/u/mr-fang-vlsi)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/mr-fang-vlsi/eda-autoidea-bridge

---


# EDA AutoIdea Bridge

## When to use
Use this skill when the user wants to combine `external_tools/autoIdea` outputs with current infra:
1. import recommended papers into local download queue,
2. merge autoIdea recommendations with landscape feedback queue,
3. ingest autoIdea idea/experiment drafts into a governed report for follow-up debate and hypothesis design.

## Inputs
- `external_tools/autoIdea` clone with at least one of:
  - `citation_graph_output/recommended_top30.json`
  - `outputs/final_idea.txt`
  - `outputs/experiment.txt`
  - `citation_graph_output/historical_problem_formulation.txt`

## Execution
Run:

```bash
python3 scripts/common/autoidea_bridge.py --autoidea-root external_tools/autoIdea
```

## Outputs
- `docs/papers/queues/autoidea_recommended_top30_bridge.tsv`
- `docs/papers/queues/literature_feedback_merged_with_autoidea.tsv`
- `docs/knowledge_base/99_AUTOIDEA_FUSION_REPORT_20260306.md`

## Integration contract
1. New autoIdea-driven queue rows must still pass current paper/validation gates.
2. Imported idea/experiment drafts are treated as candidates, not accepted plans.
3. Promotion requires `control-theory-veto` and non-regression execution evidence.