Skip to main content

Squillify development is paused indefinitely. Read why, and discover what this experiment achieved. Read the story

MicroSkills Reference

Redact Secrets

security.redact_secrets

ProductionHeuristic (Class C)Deterministicverifiablefastv1.0.0
No network

Overview

You pass text, code, config, or logs; we scan it with a curated, ReDoS-safe detector pack (cloud keys, tokens, private keys, JWTs, credentialed URLs, generic API-key assignments) and replace matches using your chosen redaction mode. We never hand back the full secret — only the placeholder we wrote and a truncated one-way SHA-256 fingerprint you can use to correlate repeats. Do not use this as compliance-grade DLP or as proof that your text is secret-free.

Secrets leak through logs, tickets, and prompts — you need a deterministic way to sanitize content before you store or transmit it.

Reliability

VERIFIABLE and deterministic for a fixed input, detector set, and mode on this deployment. The detector pack is versioned and shared with security.detect_secrets; a clean result reduces leak risk but is not certification of secret-freedom. Input validation (not retryable): missing text; text too long; unknown detector; empty detectors array; invalid redactionMode; maxFindings out of range.

Context footprint

Replaces ad-hoc LLM redaction with stable rules. A successful call costs 2 execution credits.

When to use it

Before you persist, display, or transmit free-form content that may contain credentials.

As your sole compliance or DLP control — coverage is curated and best-effort.

Scope

Note
Redacts secrets detected by deterministic, ReDoS-safe patterns from free-form text.

Non-goals

  • Does not guarantee complete secret removal (best-effort)
  • Does not validate secrets against external services
  • Does not return full secret values
  • Does not salt the fingerprint — a low-entropy secret is still guessable from it
  • NOT Gitleaks/TruffleHog parity — a curated pattern pack, not a full secret scanner
  • Known false negatives: unlisted providers, obfuscated/split secrets, and low-entropy custom tokens will pass through un-redacted

Properties

IdempotentRetry safeStreaming supportedFile input supportedRequires external service

Capabilities

analysisdata-transformation

Cost & billing

Flat rate per invocation

MetercallsBillable onsuccessCost per call2 credits

File handling

Accepts file_refProduces file_refMax input1 MB

How we handle your data

Sensitive processing
What this means

This skill is designed to detect, mask, redact, validate, or otherwise process sensitive data.

Squillify guarantees
  • Runs inside Squillify
  • This skill runs inside Squillify.
  • Payloads are not sent to an external service by this skill.
  • Squillify does not permanently store submitted payloads.
  • Telemetry records metadata only, not submitted content.
  • Submitted payloads are not used for training.
  • Submitted payloads are not reviewed by humans.
  • Targets sensitive data
  • The response may contain submitted or transformed content.
Your responsibility

Squillify does not determine whether submitted content is sensitive. Your application is responsible for deciding what data it is allowed to submit. Your application is responsible for securing the response.

Read the Microskill Data Handling Policy

Input schema

FieldTypeRequiredDescription
textstringYesFree-form text, code, or configuration to redact.
detectorsarrayNoFrom 1 through 8 named detector selectors; omit it to run every published detector.
redactionModestringNoReplacement policy; omit it for typed_placeholder.
preserveLengthbooleanNoWhether full replacement preserves length; omit it for false.
includeFindingsbooleanNoWhether receipts are returned; omit it for true.
maxFindingsintegerNoReplacement cap; omit it for 200.

Output schema

FieldTypeRequiredDescription
redactedTextstringYesBounded transformed text.
redactionCountintegerYesApplied replacement count.
findingsarrayYesFrom 0 through 1,000 redaction receipts, each a closed non-secret replacement record.

API contract

POST /v1/execute
curl
curl -sS -X POST "$BASE/v1/execute" -H "Authorization: Bearer $KEY" -H 'Content-Type: application/json' -d '{"skillKey":"security.redact_secrets","input":{"text":"aws_key = AKIAIOSFODNN7EXAMPLE"}}'
Integrations
TypeScript SDK
import { Squillify } from "@squillify/sdk";

const sq = new Squillify({ apiKey: process.env.SQUILLIFY_API_KEY });

const { output } = await sq.execute("security.redact_secrets", {"text":"aws_key = AKIAIOSFODNN7EXAMPLE"});

Errors

CodeHTTPRetryableDescription
VALIDATION_ERROR400NoThe request payload failed schema validation.
UNAUTHORIZED401NoMissing or invalid API key.
SKILL_NOT_ACCESSIBLE403NoThe API key is not granted access to this skill.
SKILL_NOT_AVAILABLE_FOR_PLAN403NoThis skill is not included in your current plan.
PLAN_ELIGIBILITY_UNAVAILABLE503YesPlan eligibility could not be decided. Retry after backoff.
SKILL_NOT_FOUND404NoThe requested skill or version does not exist.
TENANT_RATE_LIMITED429YesRequest rate limit exceeded. Retry after backoff.
QUOTA_EXCEEDED429YesPlan quota exhausted. Retryable once the quota period resets.
EXECUTION_TIMEOUT504YesExecution exceeded the allowed time budget.

Other limits

LimitValue
Recommended inline bytes32 KiB

Limitations

  • Does not guarantee complete secret removal (best-effort)
  • Does not validate secrets against external services
  • Does not return full secret values
  • Does not salt the fingerprint — a low-entropy secret is still guessable from it
  • NOT Gitleaks/TruffleHog parity — a curated pattern pack, not a full secret scanner
  • Known false negatives: unlisted providers, obfuscated/split secrets, and low-entropy custom tokens will pass through un-redacted
  • Compliance-grade DLP (best-effort only)
  • Semantic PII masking (use text.mask_pii)

How to call

POST /v1/execute (requires authentication)

Request bodyjson
{
  "skillKey": "security.redact_secrets",
  "input": {
    "text": "aws_key = AKIAIOSFODNN7EXAMPLE"
  }
}

Examples

Redact an AWS key

Inputjson
{
  "text": "aws_key = AKIAIOSFODNN7EXAMPLE"
}
Expected outputjson
{
  "redactedText": "aws_key = [REDACTED_AWS_ACCESS_KEY_ID]",
  "redactionCount": 1,
  "findings": [
    {
      "type": "aws_access_key_id",
      "start": 10,
      "confidence": 0.9,
      "fingerprint": "1a5d44a2dca19669",
      "end": 30,
      "replacement": "[REDACTED_AWS_ACCESS_KEY_ID]"
    }
  ]
}

Clean text unchanged

Inputjson
{
  "text": "nothing secret here"
}
Expected outputjson
{
  "redactedText": "nothing secret here",
  "redactionCount": 0,
  "findings": []
}
  • Sanitize logs, tickets, or prompts before storage or transmission
  • Scrub agent-generated output before display
  • Preflight documents before sharing

Avoid for

  • Compliance-grade DLP (best-effort only)
  • Semantic PII masking (use text.mask_pii)

Composable with

Safety & privacy

  • Runs without calling any external service.
  • Your payloads are never stored permanently.
  • Telemetry captures metadata only — never your payload contents.
  • We never train models on your payloads.
  • No human reviews your payloads.

Tags

securitytransformationcleanup