// Business & Operations

Name & Diacritic Integrity Auditor

Find diacritic loss, normalization drift, mixed scripts, and unsafe identity-record collisions without auto-merging people.

We earn a commission on eligible purchases after a new Agensi signup through these links. We also sell our own skills on Agensi.

Get it on Agensi Find another skill $5 one-time

Format: Agent skill, not a browser app. You need an AI agent that can load its instructions and use the required tools.

Package requirements: SKILL.md-compatible agents that can run a local Python helper. Python 3.10 or newer on Windows, macOS, or Linux using only the standard library. One documented UTF-8 JSON evidence packet; no API key or provider account.

You get: result.json with PASS, REVIEW, BLOCK, or CANNOT_ASSESS report.md human-review handoff

Licence and purchase details

Review the package licence at the destination.

Review the offer and access method at your chosen marketplace. Agensi provides connected-agent access, not a download. Buying a skill does not include an AI-agent subscription or setup service.

Agensi access: This product’s bundled-script workflow through Agensi has not been verified here. Tests of other products do not establish this package’s compatibility. Local setup examples apply to separately obtained downloadable packages.

See tested hosts, versions and workflow limits.

What problem does Name & Diacritic Integrity Auditor solve?

CRM, ticketing, payroll, event, and mailing exports strip marks or normalize names inconsistently, creating disrespectful display errors and false deduplication.

Use it to

Audit name display after a system migration
Find distinct people collapsed by accent stripping
Catch invisible controls and mixed scripts in identity data

What it returns

result.json with PASS, REVIEW, BLOCK, or CANNOT_ASSESS
report.md human-review handoff
manifest.json with stable output declarations and SHA-256 hashes

A representative input and result

fixture-backed sample

input Supply records with recordId, sourceSystem, rawName, displayName, and canonicalPersonId. Preserve source strings exactly.

result A deterministic PASS, REVIEW, BLOCK, or CANNOT_ASSESS result with stable finding codes, report, manifest, and SHA-256 receipts.

Access and approval boundaries

Terminal permission to run the bundled local Python gate.
Read access to one owner-selected UTF-8 JSON packet.
Write access only to one new owner-selected local output directory.
No browser, network, credential, account, messaging, payment, import, upload, merge, delivery, or publication permission.

Known limitations

It does not decide a person's preferred name or identity.
It never merges, deduplicates, or updates records.
Mixed scripts and normalization differences are review signals, not proof of fraud or error.

// choose with context

Is Name & Diacritic Integrity Auditor the right skill?

Best fit

CRM, ticketing, payroll, event, and mailing exports strip marks or normalize names inconsistently, creating disrespectful display errors and false deduplication.

It returns

result.json with PASS, REVIEW, BLOCK, or CANNOT_ASSESS. report.md human-review handoff.

Do not use it as

It does not decide a person's preferred name or identity.

Compatibility, access, version, and licence

Reads or accesses: Declared local scope: one owner-selected local JSON input, one new owner-selected local output directory.
Compatibility: SKILL.md-compatible agents that can run a local Python helper. Python 3.10 or newer on Windows, macOS, or Linux using only the standard library. One documented UTF-8 JSON evidence packet; no API key or provider account.
Version: 1.0.0
Licence: Review the package licence and seller terms at the linked destination.

What happens next

Open the Agensi listing, check the current offer and connect a supported agent. Agensi loads the skill through its connector; it no longer provides a download. Approve any purchase before use. Check Agensi’s current connection guide before changing an existing connection.

Questions

Does it connect to or change a live system?

No. It reads one local packet, writes one new local output directory, and makes no network request.

Does PASS prove the evidence or language is correct?

No. PASS only means no encoded rule fired in supplied normalized evidence.

What happens when evidence is missing?

The gate returns CANNOT_ASSESS or BLOCK; missing evidence never becomes a pass.

Is the result deterministic?

Yes. Findings are sorted and persisted outputs carry stable SHA-256 receipts.

Does the skill publish or make first-person claims?

No. It never publishes, contacts anyone, or adopts a founder, candidate, artist, or reviewer voice.