web-research

Neural web search and content extraction using x402-protected APIs. Better than WebSearch for deep research and WebFetch for blocked sites. USE FOR: - Deep web research and investigation - Finding similar pages to a reference URL - Extracting clean text from web pages - Scraping sites that block standard fetchers - Getting direct answers to factual questions - Research requiring multiple sources - Crawling multiple pages from a website TRIGGERS: - "research", "investigate", "deep dive", "find sources" - "similar to", "pages like", "more like this" - "scrape", "extract content from", "get the text from" - "blocked site", "can't access", "paywall" - "what is", "explain", "answer this" - "crawl", "crawl site", "scrape entire site" Use agentcash.fetch for stableenrich.dev endpoints. Prefer Exa for semantic/neural search, Firecrawl for direct scraping.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

72%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with copy-paste MCP calls and a useful endpoint table, but it inlines the full per-endpoint API reference and references two rules files that are absent from the bundle. The main research workflow also lacks validation checkpoints.

Suggestions

Create the referenced rules/getting-started.md and rules/when-to-use.md (or remove the links) so navigation targets actually exist.

Move the detailed per-endpoint options into a separate reference file, keeping only the quick-reference table and one example per endpoint in SKILL.md.

Add explicit validation/synthesis checkpoints to the Deep Research workflow (e.g., verify results are relevant before extracting, confirm sources before synthesizing).

Dimension	Reasoning	Score
Conciseness	Mostly efficient with executable examples and a tight quick-reference table, but minor padded commentary like "Semantic search that understands meaning, not just keywords" and "Great for: Finding competitor products" could be trimmed.	4 / 5
Actionability	Fully executable copy-paste-ready agentcash.fetch MCP calls with concrete URLs, methods, and bodies for every endpoint, plus documented options covering the common cases.	5 / 5
Workflow Clarity	The Deep Research checklist (search → find-similar → extract → synthesize) lacks validation checkpoints, and only the Cloudflare crawl has an explicit poll feedback loop; checkpoints are missing or implicit for the main research workflow.	3 / 5
Progressive Disclosure	Sections are reasonably organized, but the body links to rules/getting-started.md and rules/when-to-use.md which do not exist (no rules/ or references/ bundle present), and the bulk per-endpoint API reference is fully inlined rather than split out.	3 / 5
	Total	15 / 20 Passed

Description

96%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, well-structured description that clearly states capabilities and provides rich natural trigger phrases in third-person voice. The only weakness is a few overly broad triggers ("what is", "explain") that risk overlapping with general research or QA skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "Neural web search and content extraction", "Finding similar pages to a reference URL", "Extracting clean text", "Scraping sites that block standard fetchers", "Getting direct answers", "Crawling multiple pages" — giving comprehensive coverage.	5 / 5
Completeness	Explicitly answers both what ("Neural web search and content extraction using x402-protected APIs") and when (dedicated USE FOR and TRIGGERS sections with concrete trigger phrases), in third-person voice.	5 / 5
Trigger Term Quality	Comprehensive natural-term coverage with synonyms: "research", "investigate", "deep dive", "similar to", "pages like", "scrape", "blocked site", "paywall", "crawl" — phrases users would actually say.	5 / 5
Distinctiveness Conflict Risk	The x402/Exa/Firecrawl niche is distinct, but broad triggers like "what is", "explain", "answer this" and "research" overlap with general research/QA skills; not a 5 because of that minor overlap risk.	4 / 5
	Total	19 / 20 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 missing	Warning

	Total	13 / 16 Passed

Repository: Merit-Systems/agentcash-skills
Path: mcp/skills/web-research/SKILL.md
Commit: 522d5da

Reviewed: 6 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.