Jeden Skill in Manus ausführen
mit einem Klick

Jeden Skill in Manus mit einem Klick ausführen

scrub-pii-from-text

Sterne26

Forks8

Aktualisiert5. Juli 2026 um 19:44

Use this to strip PII (emails, credit cards, SSNs, phone numbers, IPs, IBANs) out of text before it is logged to a tracing backend or sent to a third-party model. Trigger on "redact PII", "scrub sensitive data", "mask PII before logging", "don't send customer data to the tracing tool", especially for finance/healthcare/regulated apps. Ships a runnable, tested scrubber with a Luhn check to cut false positives.

Installation

Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.

In Manus ausführen

Quelle

ContextJet-ai

ContextJet-ai/awesome-llm-observability

GitHub-Repository öffnen Creator-Repositorys ansehen

Download

In Manus ausführen

Verwandte BerufeSOC

Basierend auf der SOC-Berufsklassifikation

SoftwareentwicklerInformatik- und Mathematikberufe·SOC 15-1252

Datei-Explorer

3 Dateien

SKILL.md

readonly

Mehr aus diesem Repository

gleiches Repository

check-answer-consistency

ContextJet-ai/awesome-llm-observability

Use this to get a cheap, reference-free signal that an LLM answer might be made up, by sampling the same prompt a few times and measuring agreement. Trigger on "is this answer reliable", "flag low-confidence answers", "cheap hallucination check", "confidence score without a ground truth", "self-consistency check". Ships a runnable, tested scorer you can put inline or on sampled traffic.

2026-07-0526

estimate-llm-cost

ContextJet-ai/awesome-llm-observability

Use this to estimate what an LLM call or feature will cost, and to compare models on price, before or after shipping. Trigger on "how much will this cost", "estimate my OpenAI/Anthropic bill", "is a cheaper model worth it", "cost of this prompt", "project my LLM spend". Ships a runnable, tested calculator so the numbers are real, not hand-waved.

2026-07-0526

validate-genai-spans

ContextJet-ai/awesome-llm-observability

Use this to check that your LLM tracing actually emits complete, spec-compliant spans, so cost/latency/model dashboards downstream are not full of holes. Trigger on "are my traces complete", "validate my instrumentation", "my spans are missing fields", "lint my OTel GenAI spans", "test my tracing". Ships a runnable, tested validator you can drop into your instrumentation tests.

2026-07-0526

annotate-traces-for-review

ContextJet-ai/awesome-llm-observability

Use this to set up human review and annotation of LLM traces, so people (often domain experts) can label outputs, do error analysis, and build a trustworthy golden dataset. Trigger on "review my LLM outputs", "have an expert label these", "error analysis", "annotate traces", "build a golden dataset", or when automated evals are not enough for a high-stakes or specialized domain. Looking at your data is the highest-ROI thing you can do.

2026-07-0526

measure-agent-task-success

ContextJet-ai/awesome-llm-observability

Use this to measure whether an AI agent actually completed its task end to end, not just whether individual LLM calls looked fine. Trigger on "is my agent working", "measure agent success rate", "evaluate my agent", "how good is my agent", "agent completion rate", or evaluating a multi-step/tool-using agent. Score the outcome of the whole task, plus the path it took.

2026-07-0526

set-up-ab-testing

ContextJet-ai/awesome-llm-observability

Use this to test an LLM change (new prompt, new model, new retrieval) on real traffic before rolling it out to everyone. Trigger on "A/B test my prompt", "roll out a new model safely", "compare two prompts in production", "canary this change", "does this actually improve things for real users". Measure impact on real users, gated, before a full switch.

2026-07-0526

name	scrub-pii-from-text
description	Use this to strip PII (emails, credit cards, SSNs, phone numbers, IPs, IBANs) out of text before it is logged to a tracing backend or sent to a third-party model. Trigger on "redact PII", "scrub sensitive data", "mask PII before logging", "don't send customer data to the tracing tool", especially for finance/healthcare/regulated apps. Ships a runnable, tested scrubber with a Luhn check to cut false positives.
license	CC0-1.0

Scrub PII from text

Observability captures prompts and completions, which in a regulated app can mean shipping account numbers or PII to a third-party backend. This skill ships a scrubber you can run in the export path so raw values never leave your process.

Use the bundled script

scripts/pii_scrub.py is pure Python, no install needed:

from pii_scrub import scrub, find_pii

scrub("email a@b.com, card 4111 1111 1111 1111")   # -> "email <email>, card <card>"
find_pii(text)                                      # [("email", "a@b.com"), ("card", ...)]

It masks emails, credit cards (Luhn-validated, so random 16-digit order ids are not flagged), US SSNs, phone numbers, IPv4 addresses, and IBANs. Run it directly: python scripts/pii_scrub.py.

How to apply it

Redact before export, not after: call scrub() in your span processor / logging hook so no path bypasses it (see redact-pii-for-tracing for where that hook goes).
Redact both directions: user input and model output (models echo PII back).
Layer a real detector for high stakes: this is a strong default, but for finance/health, add Presidio or LLM Guard plus a domain ruleset. Regex alone misses context-dependent PII.

Validation

Run the tests: pytest skills/scrub-pii-from-text/tests/. They confirm emails/SSNs/valid cards are masked, that a Luhn-invalid number is left alone (false-positive guard), that clean text is untouched, and that find_pii reports entities.

Anti-patterns

Regex-only redaction as your whole compliance story for high-stakes data.
Masking input but not model output.
Redacting after the SDK already exported the span (mask before export).