Research
Delegate all research to a sub-agent. The primary agent receives a synthesized result, not raw search output.
When to Delegate
Delegate any research task that:
- Requires 2+ searches or source lookups
- Is exploratory (unknown territory, open-ended questions)
- Spans multiple systems (workspace KB + BQ, codebase + Jira, etc.)
- Would accumulate significant raw output inline
Single targeted lookups (one grep, one file read, one known query) can run inline.
Sub-agent Type
- Explore — read-only search across filesystem, workspace KB, grep, BQ queries. Use for most research.
- general-purpose — when research requires writing intermediate files, running scripts, or multi-step computation.
Research Protocol (Brief the Sub-agent)
The sub-agent should work through these sources in order, stopping when it has sufficient confidence:
- Workspace KB —
qmd query --no-rerank "<query>" — prior analyses, domain docs, BQ data dictionary, strategy, headcount, customer context
- Codebase — Glob/Grep/Read — source of truth for implementation details
- BigQuery —
mcp__bigquery__query — live warehouse data; always dry_run first
- Jira —
mcp__jira__jira_issues_search JQL — ticket status, project decisions, delivery context
- Web —
mcp__kagi__kagi_search_fetch (search) / mcp__kagi__kagi_extract (full-page content) — external docs, standards, anything not internal. Fall back to mcp__codex__codex only for multi-step research Kagi's two tools can't cover alone.
Sub-agent Briefing Template
Every Agent tool call for research must include:
Context: [What the overall question is and why it matters — 1-2 sentences]
Question: [The specific research question to answer]
Sources: [Which layers of the research protocol to check, in order]
Output: Synthesized summary — findings, gaps, and confidence level. Under 300 words unless the question demands more detail. No raw search output.
Output Contract
The sub-agent returns a synthesized summary. The primary agent then:
- Reads the summary (not the raw search transcripts)
- Surfaces the findings to the user
- Proceeds with the task informed by the research
If the sub-agent reports gaps or uncertainty, surface those explicitly before proceeding.
1---2name: research3description: Delegate research and context-gathering tasks to a sub-agent to protect the primary context window. Use when the user asks to "research X", "look into X", "find out about X", "gather context on X", or any investigative framing where answering requires 2+ searches or multiple sources. Also use proactively before starting substantive work when prior context is unknown. Never run research inline — always delegate.4---56# Research78Delegate all research to a sub-agent. The primary agent receives a synthesized result, not raw search output.910## When to Delegate1112Delegate any research task that:13- Requires 2+ searches or source lookups14- Is exploratory (unknown territory, open-ended questions)15- Spans multiple systems (workspace KB + BQ, codebase + Jira, etc.)16- Would accumulate significant raw output inline1718Single targeted lookups (one `grep`, one file read, one known query) can run inline.1920## Sub-agent Type2122- **Explore** — read-only search across filesystem, workspace KB, grep, BQ queries. Use for most research.23- **general-purpose** — when research requires writing intermediate files, running scripts, or multi-step computation.2425## Research Protocol (Brief the Sub-agent)2627The sub-agent should work through these sources in order, stopping when it has sufficient confidence:28291. **Workspace KB** — `qmd query --no-rerank "<query>"` — prior analyses, domain docs, BQ data dictionary, strategy, headcount, customer context302. **Codebase** — Glob/Grep/Read — source of truth for implementation details313. **BigQuery** — `mcp__bigquery__query` — live warehouse data; always `dry_run` first324. **Jira** — `mcp__jira__jira_issues_search` JQL — ticket status, project decisions, delivery context335. **Web** — `mcp__kagi__kagi_search_fetch` (search) / `mcp__kagi__kagi_extract` (full-page content) — external docs, standards, anything not internal. Fall back to `mcp__codex__codex` only for multi-step research Kagi's two tools can't cover alone.3435## Sub-agent Briefing Template3637Every Agent tool call for research must include:3839```40Context: [What the overall question is and why it matters — 1-2 sentences]41Question: [The specific research question to answer]42Sources: [Which layers of the research protocol to check, in order]43Output: Synthesized summary — findings, gaps, and confidence level. Under 300 words unless the question demands more detail. No raw search output.44```4546## Output Contract4748The sub-agent returns a synthesized summary. The primary agent then:491. Reads the summary (not the raw search transcripts)502. Surfaces the findings to the user513. Proceeds with the task informed by the research5253If the sub-agent reports gaps or uncertainty, surface those explicitly before proceeding.
lanej