research
Starts Phase 1 (Research) for a topic: reads every relevant file, optionally runs a panel of Explore teammates, and saves a research artifact to .claude/state/research/. Use when the user says 'research <topic>' or '/research', or before planning work in unfamiliar code. Research
Install
npx skills add https://github.com/domengabrovsek/agent-config/tree/main/skills/research
claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install domengabrovsek-claude@llmmart
git clone https://github.com/domengabrovsek/agent-config.git
The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole domengabrovsek/claude collection as a plugin from our marketplace. Git is the plain clone.
Skill manifest
Start Phase 1 (Research) for the following topic: $ARGUMENTS
For bigger topics, run panel mode (see rules/parallel-agents.md): spawn one Explore teammate per subsystem, run one bounded cross-check round via SendMessage so teammates flag contradictions across their areas, then pull their findings together into the artifact yourself. For narrow topics, explore solo. Read all relevant files, check git history for context, and investigate related patterns.
Save the research artifact to .claude/state/research/YYYY-MM-DD-research-<topic>.md (use today's date) containing:
- Findings: what you discovered in the codebase
- Open questions: things that need clarification
- Constraints: technical or business limitations
- Related code: key files and functions involved
Do NOT propose solutions yet. Research only.
Files (claude)
-
SKILL.md 1.2 KB
--- name: research description: "Starts Phase 1 (Research) for a topic: reads every relevant file, optionally runs a panel of Explore teammates, and saves a research artifact to .claude/state/research/. Use when the user says 'research <topic>' or '/research', or before planning work in unfamiliar code. Research only - proposes no solutions." --- Start Phase 1 (Research) for the following topic: $ARGUMENTS For bigger topics, run **panel mode** (see `rules/parallel-agents.md`): spawn one `Explore` teammate per subsystem, run one bounded cross-check round via SendMessage so teammates flag contradictions across their areas, then pull their findings together into the artifact yourself. For narrow topics, explore solo. Read all relevant files, check git history for context, and investigate related patterns. Save the research artifact to `.claude/state/research/YYYY-MM-DD-research-<topic>.md` (use today's date) containing: - **Findings**: what you discovered in the codebase - **Open questions**: things that need clarification - **Constraints**: technical or business limitations - **Related code**: key files and functions involved Do NOT propose solutions yet. Research only.
Comments (0)
Sign in to join the conversation.
Reviews (0)
No reviews yet.
No comments yet.