discover

name	discover
description	Discovery phase workflows for research interviews, literature review, data discovery, and ideation. Use when the user asks to scope a topic, review the literature, find datasets, or generate viable research designs.

Discover

Launch the discovery phase and route to the appropriate worker-critic pair based on the user's need.

Read First

• AGENTS.md
• plugins/clo-author/references/runtime-activation.md
• plugins/clo-author/references/codex-workflow.md
• plugins/clo-author/references/agent-catalog.md
• plugins/clo-author/references/critic-runtime-contract.md
• plugins/clo-author/references/domain-profile.md
• plugins/clo-author/references/logging.md

Modes

• interview / scoping
• literature review
• data discovery
• research ideation

Interview / Scoping

Use a structured conversational interview when the project is still vague.

Suggested sequence:

1. big picture and motivation
2. theoretical intuition
3. data and setting
4. identification possibilities and threats
5. expected results
6. contribution relative to the literature

Produce a research specification in quality_reports/ and calibrate the domain profile when the repo is still in template mode.

Literature Review

Dispatch:

• Librarian to collect and synthesize sources
• librarian-critic via plugins/clo-author/references/active-critics/librarian-critic.md to check coverage, recency, gaps, and scope drift

Workflow:

1. read existing specs, notes, and bibliography files
2. browse when the frontier or publication status may have changed
3. search top journals, field journals, and working-paper venues
4. follow citation chains forward and backward from the most relevant papers
5. save the synthesis under quality_reports/

If the librarian-critic finds material gaps, re-dispatch Librarian once for a targeted follow-up rather than silently accepting incomplete coverage.

Data Discovery

Dispatch:

• Explorer to build the ranked inventory
• explorer-critic via plugins/clo-author/references/active-critics/explorer-critic.md to stress-test feasibility and measurement

Each data candidate should report:

• source and access level
• key variables
• coverage
• likely unit of observation
• feasibility grade
• identification fit with the proposed question

Save the inventory in quality_reports/.

Research Ideation

Use ideation when the user wants candidate paper ideas rather than a committed project.

Outputs should include:

• a short list of viable questions
• likely treatment/outcome variation
• candidate data sources
• the next best workflow step: strategize, more discovery, or abandon

Outputs

• research spec or scoping memo in quality_reports/
• literature memo in quality_reports/
• data inventory in quality_reports/
• updated domain-profile inputs when field calibration changes

Append research-journal entries when discovery artefacts materially change project direction.