Index of every page on this site: llms.txt
All skills
Run it yourself

Audit a website for AI-discoverability and on-site-engagement problems — crawlability, JS-render gaps, missing/invalid structured data, facts locked in non-text, stale or uncorroborated facts, entity ambiguity, weak on-site orientation / no context retention — and produce a report of findings plus prioritized, actionable fixes. Use when diagnosing why a brand is missing or misrepresented in AI assistants, or why visitors who arrive don't engage.

  • 3 files
  • 5.9 KB
  • Updated 2 months ago
  • GitHub
Use this Skill: https://skilld.dev/gh/tejasshukla2007/brand-ai-readiness-audit/audit-orchestrator

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈118 tokens for metadata: the name and description. ≈439 when used: this file. ≈669 more on demand in 1 file.

Description uses 5.9% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈470 characters, or 5.9%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Audit Orchestrator

When to use

  • Use this skill when you need a complete audit of a brand's visibility in AI assistants.
  • Invoke this to get a consolidated report covering both AI discoverability (off-site citations) and on-site engagement.
  • This skill orchestrates the entire audit workflow. It performs core checks directly and can also invoke subordinate skills if they exist in the marketplace.

Inputs

  • url: The target website URL or domain string to evaluate.

Procedure

  1. Parse the input url to ensure it is a valid, reachable target for the audit.
<!--Update the sub-skill list below as new skills are added to the marketplace.-->
  1. Invoke each available sub-skill on the url and collect their findings arrays. Registered sub-skills: freshness-corroboration, semantic-structure-audit, a11y-extractor-audit, context-retention-audit, content-quality-audit.
  2. Aggregate all findings from the executed sub-skills into a single, unified list. Deduplicate any overlapping issues.
  3. Rank the aggregated findings by severity (critical, high, medium, low). Ensure every finding explicitly states the evidence and includes a proactive, mechanism-sound suggested_action.
  4. Execute python scripts/compile_report.py <url> to generate the base JSON structure and automatically calculate the summary metrics (total_findings, critical, high, medium, low) based on the aggregated list.
  5. Inject the ranked findings array into the base JSON structure.

Output

Emit a single JSON report object strictly conforming to the schema defined in references/output_schema.json

The final output must be recommend-only. Do not attempt to make any changes to the site.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at 1b012a3. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Steadyupdated 2 months ago

README badge

README badge for tejasshukla2007/brand-ai-readiness-audit/audit-orchestrator