Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/winsenlabs/platos/parallel_webnpx skills add winsenlabs/platos --skill parallel_webgit clone --depth 1 https://github.com/winsenlabs/platosWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/winsenlabs/platos/parallel_web)<a href="https://agentmods.dev/skills/winsenlabs/platos/parallel_web"><img src="https://agentmods.dev/badge/skills/winsenlabs/platos/parallel_web.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00037 | $0.01504 |
| Opus 5 | $0.00018 | $0.00752 |
| Sonnet 5 | $0.00007 | $0.00301 |
| Haiku 4.5 | $0.00004 | $0.00150 |
Grade A, and why
Parallel Web Research scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
You have access to Parallel.ai — a professional-grade web research stack. Pick the right tool for the job:
parallel_search — fast, LLM-optimized web search.
- Use for: recent events, fresh facts, quick lookups where a few ranked excerpts are enough.
- Pass an
objective(the question driving the search) + up to 10 concretesearchQueries.
parallel_extract — URL → clean markdown + focused excerpts.
- Use when: you already have URLs (from
parallel_searchor the user) and need the full content, or the page is JS-rendered / a PDF. - Up to 5 URLs per call.
parallel_deep_research — multi-minute research runs.
- Use when: the question requires multi-hop reasoning, cross-source synthesis, or a specific structured output.
- Waits up to 5 minutes inline. If still running, returns
{ runId, status: "running" }— callparallel_deep_research_resultwith the runId to check back. - Choose a processor:
lite<base<core<pro<ultra<ultra8x(cost + depth scale up).
parallel_deep_research_result — poll an in-progress deep research run.
- Pass the
runIdfrom a previousparallel_deep_researchcall that returnedstatus: "running". - Returns the full result when done;
{ status: "running" }if still in progress.
parallel_findall — structured datasets.
- Use when: the user wants a list of entities (companies, products, papers…) matching criteria, with a consistent schema per row.
- Supply a JSON Schema describing one row; Parallel returns an array of rows matching it.
parallel_monitor_create — persistent web monitors.
- Use when: the user asks to be notified of changes on a page ("tell me when X launches", "watch this pricing page").
- Returns a monitor id. Optionally posts to a webhook when the monitor fires.
Guidelines:
- Always cite URLs returned by these tools with
[1](url)footnotes — never fabricate citations. - Prefer
parallel_searchfor single-turn factual questions; escalate toparallel_deep_researchonly when the question genuinely needs it (latency + cost are higher). - For
parallel_extract, keep to URLs returned byparallel_searchor explicitly given by the user.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today First seen · 73 lines · 0 tokens per session scan A 9c4e9547903f
Parallel Web Research is a skill published in the GitHub repository winsenlabs/platos (24 stars, last pushed yesterday), licensed Apache-2.0. It adds 37 tokens to every session and 1,504 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other skills, from other repositories
cn-check
Install and run the Continue CLI (cn) to execute AI agent checks on local code changes. Use when asked to "run checks", "lint with AI", "review my changes with cn", or set up Continue CI locally.
docs-style
Style guidelines for writing and updating documentation. Use when writing new docs, updating existing docs, or reviewing docs for quality.
bio-ortholog-inference
Pull pre-computed ortholog calls from public databases (OrthoDB, Ensembl Compara, OMA browser, eggNOG, PANTHER, KEGG Orthology, HomoloGene) via their REST APIs. Use when orthologs are already curated upstream, when the question is "what is the X ortholog of Y" rather than "how to infer orthology de novo", when…
bio-differential-expression-batch-correction
Handles batch effects in bulk RNA-seq via design-matrix inclusion (the correct path for DE), ComBat/ComBat-seq for visualization, SVA for unknown latent factors, RUVSeq for negative-control-gene-anchored unwanted variation, and limma::removeBatchEffect for plotting only. Encodes the Nygaard 2016 cardinal sin against…
admet_genetic
ADMET-guided genetic molecule optimization workflow from seed SMILES; use when the agent needs to build or run an RDKit/SA-Score/ADMET-AI GA pipeline for molecule optimization, enforce molecule lineage logs, render optimization-history HTML dashboards, and write candidate triage reports.
bioprobench
Score an LLM's biological-protocol reasoning on the BioProBench benchmark: protocol QA, step ordering, error detection, protocol generation, and LLM-judged error reasoning; or generate the responses.