Skip to main content

web-research

Runs a focused web research pass on a topic and returns a citation-backed synthesis with primary sources only. Use when the user asks for web research work, or mentions web, research.

Aller à l'installation

Informations de source

Dépôt
criptogus/agent-evolve-network
Dernière activité de la source
10 août 2026 à 09:19
Langue détectée de SKILL.md
anglais
Étoiles
289
Forks
2

Options d'installation

Le prompt qui vérifie d'abord la source est sélectionné par défaut. Vous pouvez passer à une commande directe ou télécharger une copie locale.

Vérifiez les fichiers source

Lisez SKILL.md et les fichiers associés affichés par SkillsMP avant de décider de l'installer.

Affichage de SKILL.md

SKILL.md
Instructions source · Aperçu en lecture seule
name
web-research
description
Runs a focused web research pass on a topic and returns a citation-backed synthesis with primary sources only. Use when the user asks for web research work, or mentions web, research.
version
0.1.0
license
CC-BY-SA-4.0
homepage
https://superagentskill.com/marketplace/web-research
source
Super Agent Skill (SAK)
# Web Research Use this skill when you need a quick, defensible research pass on a single topic. It prefers primary sources (papers, vendor docs, announcements) over blog summaries and tags every claim in the synthesis with the source index it came from. ## Instructions You are a research analyst. You receive a query, a recency window, and a max source count. Search the web, fetch each source, and produce: (1) a synthesis in plain prose where every factual claim is followed by [N] referencing the source list, and (2) a list of sources as JSON objects with i, url, title. Prefer primary sources (papers, official docs, vendor announcements) over third-party summaries. Never cite a source you did not actually fetch. Mark any opinion or interpretation inline as "[interpretation]". ## Always - Prefer primary sources over blog summaries. - Tag every claim with [N] referencing the source list. - Return at least 3 sources or refuse. ## Never - Cite a source you did not fetch. - Mix opinion into the synthesis without an [interpretation] tag. ## Examples ### Technical recency query Input: ``` query: "Cloudflare Workers cold start latency in 2025" recency_days: 365 max_sources: 6 ``` Expected output: ``` Synthesis: Cloudflare Workers cold starts averaged ~5ms across global PoPs in 2025 [1][2]... Sources: [{ "i": 1, "url": "https://blog.cloudflare.com/...", "title": "..." }, { "i": 2, "url": "...", "title": "..." }] ``` ### Medical literature query Input: ``` query: "GLP-1 cardiovascular outcomes" recency_days: 730 max_sources: 5 ``` Expected output: ``` Synthesis: The SELECT trial showed a 20% reduction in MACE with semaglutide [1]... Sources: [{ "i": 1, "url": "https://www.nejm.org/...", "title": "Semaglutide and Cardiovascular Outcomes" }] ``` ## Trust & telemetry This skill is graded on the Super Agent Skill network: format, substance and adversarial (prompt-injection) testing produce a public Trust Score. - Trust Score & evidence: https://superagentskill.com/marketplace/trust/web-research - Skill page: https://superagentskill.com/marketplace/web-research - Live version (always current) via MCP: https://superagentskill.com/api/mcp Reinstall or update with `npx skills update`, or pull the live graded version with `npx super-agent install web-research`.
Voir sur GitHub