agentsclimarketplace

Finding

Skill Crawlio-app/crawlio-plugin/skills/finding

AI skills for website crawling, observation, and analysis — powered by Crawlio

Install
npx -y skills add Crawlio-app/crawlio-plugin --skill finding

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use this skill when the user asks to "create a finding", "record an insight", "what findings exist", "show findings", or wants to create or review evidence-backed analysis insights from crawl observations. Creates and queries curated findings with evidence chains.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

4.0 KB, as published. Nobody here has run it

finding

Create and query curated findings in Crawlio's observation log. Findings are agent-created insights backed by observation evidence.

When to Use

Use this skill when the user wants to:

  • Record an insight or issue discovered during analysis
  • Create an evidence-backed finding that persists across sessions
  • Review previously created findings for a site

Creating Findings

Findings are the agent's judgment layer on top of raw observations. A good finding:

  1. Has a clear, descriptive title
  2. References specific observation IDs as evidence
  3. Includes a synthesis explaining the pattern or issue

Workflow

  1. Query observations to identify patterns:
get_observations({ host: "example.com", source: "extension", limit: 50 })
  1. Identify the pattern — look for recurring issues, framework signals, error patterns, or notable behaviors.

  2. Create the finding with evidence:

create_finding({
  title: "Mixed content: HTTP images on HTTPS page",
  url: "https://example.com",
  evidence: ["obs_a3f7b2c1", "obs_b4e8c3d2"],
  synthesis: "Homepage loads 3 images over HTTP despite serving over HTTPS. Network observations show requests to http://cdn.example.com/img/ which should use HTTPS. This triggers mixed content warnings in Chrome and may cause images to be blocked in strict mode.",
  confidence: "high",
  category: "security"
})

Parameters

ParameterTypeRequiredDescription
titlestringYesShort, descriptive title
urlstringNoURL this finding relates to
evidence[string]NoArray of observation IDs (obs_xxx)
synthesisstringNoDetailed explanation
confidencestringNohigh, medium, low, or none
categorystringNoDimension (e.g. performance, security, framework)

Finding Quality Checklist

  • Title: Is it specific? "Mixed content on homepage" > "Issue found"
  • Evidence: Do the observation IDs actually support the claim?
  • Synthesis: Does it explain why this matters, not just what was observed?
  • URL: Is it scoped to the right page or left empty for site-wide findings?

Querying Findings

All Findings

get_findings({})

Findings for a Specific Host

get_findings({ host: "example.com" })

Recent Findings

get_findings({ limit: 10 })

Finding Categories

When creating findings, consider these common categories:

CategoryExample Title
Performance"Render-blocking scripts delay FCP by 2.3s"
Security"Mixed content: HTTP resources on HTTPS page"
SEO"Missing meta descriptions on 12 pages"
Framework"Next.js App Router with ISR detected"
Errors"3 JavaScript errors on product pages"
Structure"Orphaned pages not linked from navigation"
Accessibility"Missing alt attributes on hero images"

Evidence Chain

The full evidence chain workflow:

  1. analyze_page → returns evidenceId
  2. create_finding → reference the evidenceId in the evidence array
  3. get_observation → verify the evidence entry exists and supports the finding

Tips

  • Create findings as you analyze, not all at the end — they persist across sessions
  • Reference multiple observation IDs when a finding draws from several data points
  • Use synthesis to explain the impact, not just restate the observation
  • Findings with evidence chains are much more useful than findings without
  • Use confidence to signal how strongly the evidence supports the claim
  • Use category to enable filtering by dimension (performance, security, SEO, etc.)

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.