Index of every page on this site: llms.txt
All skills
tejasshukla2007 avatar

/freshness-corroboration

@1b012a3
Run it yourself

Evaluates temporal freshness signals, external trust-claim corroboration, and entity disambiguation on a target domain. Detects stale copyright dates, outdated content publication/modification dates, uncorroborated performance claims, unattributed testimonials, and absence of any freshness indicators.

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

Nothing lands on disk. Nothing to clean up.

Fork this Skill

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

SKILL.md

≈82 tokens for metadata: the name and description. ≈750 when used: this file. ≈189 more on demand in 1 file.

Description uses 4.1% 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 ≈328 characters, or 4.1%.

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.

Freshness and Corroboration Audit

When to use

Execute this skill when the audit orchestrator requests an evaluation of a domain's content freshness, claim corroboration, and entity clarity for AI discoverability.

Inputs

  • target_url: The full URL or domain name to be audited.

Procedure

  1. Fetch the raw HTML content of the target_url homepage.
  2. Execute python scripts/scan_signals.py <url> passing the full HTML via stdin. The script strips <script> and <style> blocks, then runs deterministic regex-based checks to extract boolean flags and evidence snippets.
  3. Read the Tool Output Data. You are strictly forbidden from hallucinating evidence. You must only evaluate the flags and snippets explicitly provided by the script.
  4. Ignore all Semantic Structure signals. Do not report on JSON-LD, Schema.org, or /llms.txt.
  5. Map the extracted evidence to the Deterministic Grading Rubric below. If a flag is false in the tool output, do not generate that finding.
  6. For each finding, use the corresponding snippet field from the script output as the basis for the evidence string. Do not fabricate evidence beyond what the script provides.
  7. Sort all generated findings alphabetically by their id.
  8. Format all detected issues strictly into the JSON structure defined in references/finding_schema.json.

Deterministic Grading Rubric

You must assign severities and generate findings strictly according to this matrix.

Evaluation Rubric

Map the output flags EXACTLY to the following Finding IDs. If html_fetch_failed is true, you MUST emit ONLY FC-000 and omit all other findings. Note: For FC-000, phrase the evidence and suggested_action as an informative warning. Inform the user that the audit was inconclusive due to a firewall, and gently recommend verifying that official AI crawlers (like GPTBot) are whitelisted.

Python Output Flag Finding ID Title Severity Default Priority
html_fetch_failed: true FC-000 Audit Inconclusive: Bot Protection Active medium medium
outdated_copyright_found: true FC-001 Outdated Copyright Year Signals Stale Content high high
uncorroborated_claims_found: true FC-002 Uncorroborated Performance Claims Without Source Links medium medium
unattributed_quotes_found: true FC-003 Unattributed Testimonials Lacking Entity Attribution medium medium
stale_dates_found: true FC-004 Stale Content Dates Exceeding 5-Year Freshness Threshold high high
no_freshness_signals: true FC-005 No Temporal Freshness Signals Detected on Page medium medium
social_proof_unlinked_found: true FC-006 Unlinked Social Proof Claims Lacking Verifiable Source medium medium
stale_cdn_cache_found: true FC-007 Stale CDN Cache Detected via HTTP Age Header low low

Output

Emit only a valid JSON array matching the exact structure dictated in references/finding_schema.json.

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/freshness-corroboration