More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    🔒 PII & Data-Leak Scanner

    2

    Get severity-ranked file and line findings without sending data to an external service.

    $15

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    3 usesSecurity scanned
    🔒 PII & Data-Leak Scanner

    🔒 PII & Data-Leak Scanner

    Example session with this skill installed

    Scan the SQL, CSV, JSON, and log files under tests/fixtures and tmp/logs for likely PII. Do not write files or reproduce matched personal-data values in the report.

    • Read your context and instructions
    • Compiled the pii & data-leak

    PII Data Leak Scanner Report

    Files scanned: 4
    Findings: 15

    [HIGH] PII002 - tmp/app.log:1

    Email-looking value found.
    Evidence: <redacted-email>
    Remediation: Replace real-looking PII with synthetic or anonymized values.

    [CRITICAL] PII004 - tmp/app.log:1

    Credit-card-like number found.
    Evidence: <redacted-card>
    Remediation: Replace real-looking PII with synthetic or anonymized values.

    [HIGH] PII001 - fixtures/profile.json:3

    PII-indicating JSON key found.
    Evidence: phone
    Remediation: Rename, tokenize, hash, or remove PII keys from shared fixtures.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Catch personal data before code or logs are shared

    Scan local SQL, CSV, JSON, seed, fixture, and log files for PII-indicating column or key names and email-, SSN-, card-, and phone-like values. Get severity-ranked file and line findings without sending data to an external service.

    Keep detected values out of the report

    The v1.1.1 scanner replaces matched email-, SSN-, card-, and phone-like values with redacted placeholders and limits structural evidence to the matched key or column name. Reports preserve the location, rule, and remediation without copying nearby record content.

    Run locally and read-only

    The bundled scanner uses only the Python 3 standard library. It requires no third-party package, network access, API key, environment variable, database connection, secret read, or file write.

    Know what it can and cannot conclude

    The scanner uses regex and keyword heuristics, so false positives and false negatives are possible. It covers common PII indicators and US-style SSN, card, and phone patterns; it is not a DLP system, legal advice, or proof of GDPR, CCPA, or SOC 2 compliance.

    Review evidence before remediation

    Confirm whether a match is real personal data, synthetic test data, an expected schema field, or a harmless format match. The included checklist helps reviewers document context, minimization, retention, access, and the next human decision.

    Check coverage before calling a scan complete

    Version 1.1.1 reports missing, unreadable, invalid-encoding, linked and explicitly unsupported inputs in both formats. Partial or zero supported-file coverage, and failed report output, return exit 2. A completed scan returns 0 even when findings exist: read findings and assessment_status. Matched excerpts are redacted; paths and filenames remain visible. Markdown escapes path formatting; JSON retains original path values. This is neither a full anonymizer nor a compliance certification.

    Verified workflow

    The v1.1.1 creator-delivered files were retrieved through Agensi and tested in Claude on Linux in an isolated folder. The bundled suite ran 37 tests: 33 passed and four Windows-only junction checks were skipped. Local Windows stdin checks also passed. This verifies the tested creator workflow, not paid-buyer checkout or every compatible agent. The bundled regression suite requires its complete file layout.

    • Heuristic regex and keyword detection can produce false positives and false negatives.
    • File coverage is limited to .sql, .csv, .json, and .log, including those inside seed and fixture folders.
    • SSN, card, and phone rules focus on common US-style patterns; locale-specific or unusual formats may be missed.
    • Key and column names can indicate expected schema fields rather than an actual leak.
    • The report is a review aid, not a DLP system, legal advice, or proof of GDPR, CCPA, or SOC 2 compliance.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    3 uses

    Times used to date

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed today

    Permissions

    Terminal / Shell
    Read Files

    File Scopes

    pii-data-leak-scanner/**
    **/*.sql
    **/*.csv
    **/*.json
    **/*.log

    All scanning is local. The bundled scanner reads matching files, prints markdown or JSON findings, and never transmits, uploads, or stores your data anywhere. It installs nothing and modifies nothing. Any write, install, deploy, or live-account action requires explicit user confirmation.

    Listed5 months ago
    Updatedtoday

    What's inside

    Frequently Asked Questions

    It runs a deterministic local scanner across .sql, .csv, .json, and .log files, reports the file, line, rule, and remediation, and redacts matched email-, SSN-, card-, and phone-like values from evidence. Results remain heuristic and require human review.

    It works with Claude Code, Codex CLI, Cursor, OpenCode/OpenClaw, Gemini CLI, and other agents that load SKILL.md folders and can read the scoped files. Python 3 runs the bundled scanner; checklist mode remains available without it.

    No. The scanner uses only the Python 3 standard library and needs no API key, external service, network access, database connection, environment variable, or third-party security package.

    The package includes the SKILL.md workflow, read-only scanner, clean and risky fixtures, scanner tests, an audit checklist, and remediation references. Version 1.1 redacts matched personal-data values in report evidence by default.

    No. Regex and keyword heuristics can miss unusual formats or flag harmless examples. The report supports developer review but is not a DLP system, legal advice, or proof of GDPR, CCPA, SOC 2, or other compliance.

    New in Security