From harness-engineering
Track 2 Phase 1: Research a topic using parallel sub-agents and PTC scripts
npx claudepluginhub emingenc/harness-engineering --plugin harness-engineering# /research — Track 2 Research Phase You have been asked to research: **$1** Follow the researcher skill workflow exactly: 1. **Define** 3-5 specific research questions about "$1" 2. **Local search** first via PTC: 3. **Read** only the most relevant files from the search results 4. **Sub-agents** for deeper questions (spawn in parallel if independent): Use Agent tool with `subagent_type: "Explore"` for each independent question 5. **External docs** via context7 MCP only if local search is insufficient 6. **Synthesize** findings: 7. **Present** key findings and recommend ...
/researchPerforms adaptive deep web research on a query with configurable --depth and --strategy options. Outputs markdown report with executive summary, analysis, confidence scores, and cited sources.
/researchPrompts for research intensity (quick/standard/deep) then invokes octo:discover skill for multi-AI orchestration, synthesis, and analysis.
/researchConducts multi-turn deep research on a codebase topic over 5 iterations, tracing code paths with citations, Mermaid diagrams, tables, and confidence ratings.
/researchConducts institutional-grade equity research analysis on a stock ticker via web searches, producing a structured report with summary, financials, catalysts, valuation, risks, and technicals.
/researchGathers knowledge from trusted web sources and files, cross-references claims across multiple sources, and produces cited research documents in docs/research/. Optional skill distillation via --skill-for.
/researchRuns research phase for current or specified spec: gathers context, optional interview, parallel subagent research, synthesizes research.md, reviews, approves, and finalizes state.
You have been asked to research: $1
Follow the researcher skill workflow exactly:
Define 3-5 specific research questions about "$1"
Local search first via PTC:
python3 ${CLAUDE_PLUGIN_ROOT}/skills/researcher/scripts/search_local.py <relevant_terms>
Read only the most relevant files from the search results
Sub-agents for deeper questions (spawn in parallel if independent):
Use Agent tool with subagent_type: "Explore" for each independent question
External docs via context7 MCP only if local search is insufficient
Synthesize findings:
python3 ${CLAUDE_PLUGIN_ROOT}/skills/researcher/scripts/format_findings.py \
--topic "$1" --output "workspace/research/$1.md" --findings '<json>'
Present key findings and recommend next steps
Log:
python3 ${CLAUDE_PLUGIN_ROOT}/scripts/progress.py append "Research complete: $1"