Parallel Deep Research
Run exhaustive, citation-rich web research via parallel-cli when a solo builder explicitly wants a comprehensive report—not a quick lookup.
Overview
parallel-deep-research is an agent skill for the Idea phase that runs Parallel’s CLI deep-research processor to produce exhaustive, downloadable web research reports when you explicitly ask for deep or comprehensive inve
Install
npx skills add https://github.com/parallel-web/parallel-agent-skills --skill parallel-deep-researchWhat is this skill?
- Starts a parallel-cli research run with topic arguments and a reusable lowercase hyphenated output basename
- Polls job status with --timeout 600s until completed or failed, then surfaces error text on failure
- Downloads bundled markdown (and optional extras) with parallel-cli research download to the run output directory
- Hard gate: only for explicit deep/exhaustive/comprehensive/thorough wording—otherwise parallel-web-search
- Requires parallel-cli ≥ 0.3.0 with pipx/parallel-cli update path when CLI flags mismatch
- 10–100x slower and more expensive than parallel-web-search per skill guidance
- Three-step workflow: research run, status poll with 600s timeout, research download
- Requires parallel-cli version 0.3.0 or newer
Adoption & trust: 9.1k installs on skills.sh; 56 GitHub stars; 1/3 security scanners passed (skills.sh audits).
What problem does it solve?
You need a thorough, sourced research report on a topic, but ad-hoc web search and quick agent lookups are too shallow for decisions you are about to commit to.
Who is it for?
Solo builders who explicitly request exhaustive research on markets, competitors, or technical choices and already have parallel-cli installed with network access.
Skip if: Normal research, quick facts, or fact-checking—use parallel-web-search instead unless the user literally asks for deep, exhaustive, comprehensive, or thorough investigation.
When should I use this skill?
User explicitly says deep research, exhaustive, comprehensive report, or thorough investigation; not for normal research or lookup requests.
What do I get? / Deliverables
You get a completed parallel-cli research job with markdown (and related artifacts) saved under a topic-named directory, ready to cite in specs, validation notes, or launch content.
- Completed research job with downloaded markdown report in a topic-named directory
- Optional continuation context via --previous-interaction-id for follow-up runs
- Actionable error surfacing when the research run fails
Recommended Skills
Journey fit
Deep research is a canonical Idea-phase capability for opportunity and landscape investigation before committing to build; the skill is gated to explicit exhaustive requests. The workflow produces long-form, sourced reports on topics, competitors, and markets—the core output of the research subphase shelf.
How it compares
Use for explicit deep-research CLI runs with polling and file download; use parallel-web-search for everyday lookups that should stay fast and cheap.
Common Questions / FAQ
Who is parallel-deep-research for?
Indie and solo builders using agent coding environments who want file-backed, comprehensive web research through parallel-cli when scoping ideas, competitors, or markets.
When should I use parallel-deep-research?
Use it in Idea (research) when you say deep research, exhaustive, comprehensive report, or thorough investigation; optionally in Validate (scope) or Launch (geo) for the same explicit depth—not for routine lookups.
Is parallel-deep-research safe to install?
It invokes parallel-cli over the network via Bash; review the Security Audits panel on this Prism page and treat downloaded web content as untrusted until you verify sources.
SKILL.md
READMESKILL.md - Parallel Deep Research
# Deep Research Research topic: $ARGUMENTS > Requires `parallel-cli` ≥ 0.3.0. If any command below errors with `no such option`, `no such command`, or `unrecognized arguments`, the user is on an older CLI. Tell them to run `parallel-cli update` (or `pipx upgrade parallel-web-tools` if installed via pipx), then retry. ## When to use (vs parallel-web-search) ONLY use this skill when the user explicitly requests deep/exhaustive research. Deep research is 10-100x slower and more expensive than parallel-web-search. For normal "research X" requests, quick lookups, or fact-checking, use **parallel-web-search** instead. ## Step 1: Start the research Choose a descriptive filename based on the topic (e.g., `ai-chip-market-2026`, `react-vs-vue-comparison`). Use lowercase with hyphens, no spaces. Reuse this base name in step 2 as `-o "$FILENAME"`. ```bash parallel-cli research run "$ARGUMENTS" --processor pro-fast --text --no-wait --json ``` The `--text` flag tells the API to return a markdown report (with inline citations) when the task completes, instead of the default structured JSON. Use it for narrative/report-style requests, which is what most users want from "deep research." Drop `--text` if the user explicitly wants structured JSON output. Optional with `--text`: pass `--text-description "Keep under 1500 words, focus on M&A activity"` to steer length, format, or focus. If this is a **follow-up** to a previous research or enrichment task where you know the `interaction_id`, add context chaining: ```bash parallel-cli research run "$ARGUMENTS" --processor lite-fast --text --no-wait --json --previous-interaction-id "$INTERACTION_ID" ``` By chaining `interaction_id` values across requests, each follow-up question automatically has the full context of prior turns — so you can drill deeper without restating what was already researched. Use a lighter processor (`lite-fast` or `base-fast`) for follow-ups since the heavy lifting was done in the initial turn. This returns instantly. Do NOT omit `--no-wait` — without it the command blocks for minutes and will time out. Processor options (choose based on user request): | Processor | Expected latency | Use when | |-----------|-----------------|----------| | `lite-fast` | 10–60s | Quick lookups, follow-ups | | `base-fast` | 15–100s | Simple questions | | `core-fast` | 1–5 min | Moderate research | | `pro-fast` | 2–10 min | **Default** — exploratory research, good depth/speed balance | | `ultra-fast` | 5–25 min | Multi-source deep research (~2× cost) | | `ultra2x-fast` / `ultra4x-fast` / `ultra8x-fast` | up to 2 hr | Hardest questions, only when explicitly requested | Notes on the `-fast` suffix: `-fast` tiers use cached web data and are quicker. The non-fast variants (`pro`, `ultra`, etc.) re-fetch fresher data — slower but better for very recent events. Default to `-fast` unless the user specifically asks about news from the last day or two. Run `parallel-cli research processors` to see the full list with latencies. Parse the JSON output to extract the `run_id`, `interaction_id`, and monitoring URL. Immediately tell the user: - Deep research has been kicked off - The expected latency for the processor tier chosen (from the table above) - The monitoring URL where they can track progress Tell them they can background the polling step to continue working while it runs. ## Step 2: Poll for results ```bas