claude-deep-research

name	claude-deep-research
description	Run a Claude Code dynamic deep-research workflow on a question (multi-agent fan-out with WebSearch + cross-checked citations).

You are the Codex skill wrapper for the cc dispatcher's deep-research subcommand.

Resolve <plugin-root> as the parent directory of the skills/ directory that contains this file (so <plugin-root>/scripts/cc.mjs is the dispatcher). Confirm <plugin-root>/scripts/cc.mjs exists before running.

Run:

node "<plugin-root>/scripts/cc.mjs" deep-research -- "<question>"

Return the dispatcher's stdout verbatim. If the command exits non-zero, show stderr/stdout to the user and explain that the dispatcher failed. Do not reimplement the command logic yourself.

Behavior rules:

• Treat the user's remaining text after the skill invocation as the research question. If empty, ask the user for a question before running.
• Forward only these flags when the user explicitly requests them: --name, --model, --effort, --permission-mode, --add-dir, --bypass-permissions, --dangerously-skip-permissions, --allow-dangerously-skip-permissions, --mcp-config, --agent, --agents, --allowedTools, --allowed-tools, --disallowedTools, --disallowed-tools, --tools, --settings, --setting-sources, --strict-mcp-config, --append-system-prompt, --system-prompt, --plugin-dir, --plugin-url, --bare, --safe-mode, --ide, --chrome, --no-chrome, --disable-slash-commands, --exclude-dynamic-system-prompt-sections, --verbose, --json.
• The --effort flag accepts low, medium, high, xhigh, or max (Claude CLI valid set). The ultracode value is TUI-only and is silently ignored when passed via --effort. To trigger Claude Code's auto-orchestration workflow planning, use $claude-workflow instead — it injects the ultracode: keyword that activates the same behavior.
• The user may also pass --yes to skip the first-run privacy acknowledgement. Do not inject --yes for ordinary jobs. Exception: if the user explicitly asks for trusted unattended Claude work and you forward --bypass-permissions, --dangerously-skip-permissions, or --permission-mode bypassPermissions, include --yes in the same command so the job stays inside the cc status/result surface instead of falling back to native claude. If the dispatcher reports that an acknowledgement is required in any other case, surface that message to the user instead of retrying with --yes.
• If the user explicitly asks for trusted unattended Claude work, or has already approved that mode for the current task/session/project, forward --bypass-permissions for fresh local shell/tool automation jobs. The dispatcher translates bypass aliases to Claude Code's literal --dangerously-skip-permissions flag. If a bypass-launched job still needs interactive input immediately, the dispatcher exits non-zero and marks the job failed instead of returning a blocked worker. This does not resolve browser selection, passkeys, or other local user-gesture prompts.
• Do NOT forward --allow-edit — it is not applicable to this subcommand.

Approval flow — important:

This skill starts a Claude Code background session with /deep-research <question> injected as the opening slash command. The /deep-research slash command triggers Claude's bundled workflow runtime, which spawns multiple agents fanning out web searches, fetching sources, adversarially verifying claims, and synthesizing a cited report. Current Claude Code versions may present a dynamic workflow approval gate before subagents start. After the job ID and Claude session short ID are printed, the user can run:

claude attach <shortId>

using the printed Claude session short ID to watch progress or approve the gate. The --yes flag only acknowledges the plugin privacy prompt; it does not approve Claude Code workflow gates.

WebSearch requirement: the /deep-research workflow requires the WebSearch tool. This tool is auto-available in standard Claude Code background sessions (confirmed via Plan 0013 OQ-B empirical probe).

Cost notice:

Research-grade workflows use significant tokens. The /deep-research runtime can spawn multiple agents fanning out parallel web searches, per workflow limits of 16 concurrent agents and 1000 total agents per run. Recommend using narrow, specific questions over broad sweeps to contain cost. Use $claude-stop to terminate a deep-research session early.

When deep-research stalls

If $claude-status <jobId> returns needs_input on the first poll and stays there, the /deep-research injection may have failed to reach the model. Stop the stalled session with $claude-stop <jobId> and re-run $claude-deep-research with the same question.

Next steps

Deep-research sessions appear as standard background jobs; after starting one:

• $claude-status — check live progress
• $claude-result — read the final cited report once the session completes
• $claude-followup — send an additional instruction mid-run
• $claude-stop — terminate the deep-research session early