26 Unix tools. One binary. Zero dependencies. · the missing coreutils for AI pipelines · vrk mcp - expose all 26 tools to any AI agent · brew install vrk - ready in 5 seconds · 26 Unix tools. One binary. Zero dependencies. · the missing coreutils for AI pipelines · vrk mcp - expose all 26 tools to any AI agent · brew install vrk - ready in 5 seconds

vrk mask

About

Redacts secrets from text before it reaches an LLM or a log file. Catches Bearer tokens, passwords, API keys, and other credentials using built-in patterns. Also flags high-entropy strings that look like secrets but don’t match a known pattern. Streams input, so it handles large files safely.

The problem

You pipe log output to an LLM for analysis and accidentally send API keys, Bearer tokens, and passwords in the prompt. The LLM provider now has your production credentials in their training pipeline.

Before and after

Before

cat output.log | sed 's/Bearer [^ ]*/Bearer ***/g' | \
  sed 's/password=[^ ]*/password=***/g'
# misses API keys, high-entropy secrets, custom patterns

After

cat output.log | vrk mask

Example

cat output.txt | vrk mask

Exit codes

CodeMeaning
0All lines processed
1Stdin scanner error or write failure
2Interactive TTY with no piped input, invalid regex

Flags

FlagShortTypeDescription
--pattern[]stringAdditional Go regex to match and redact (repeatable)
--entropyfloat64Shannon entropy threshold; lower catches more
--json-jboolAppend metadata trailer after output
--quiet-qboolSuppress stderr output