TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: Nosey Parker Explorer, a TUI for interactive review of secret exposures

3 pointsby SnowflakeOnIce3 months ago
Nosey Parker Explorer is an interactive TUI tool for reviewing possible exposed secrets detected by Nosey Parker [1], a fast secrets detector designed for offensive security (e.g., red team engagements). It makes it feasible for one person to triage thousands of findings in a few hours.<p>How did Nosey Parker Explorer come about? 2-3 years ago I was working on training ML models for tasks related to hardcoded secrets, such as TP&#x2F;FP classification and detection. These are pretty specific tasks and there are no open-access datasets. So, I had to build my own dataset of several thousand labeled examples.<p>None of the existing tools for data labeling that I ran across was a good fit for the task. They were all either too general-purpose, too complicated to set up, or too slow. So I built my own proof-of-concept purpose-built terminal-based labeling app using the excellent Textual TUI framework [2] and DuckDB [3] to build fast faceted search. A couple weeks later I had Nosey Parker Explorer.<p>Nosey Parker Explorer proved very effective at the dataset construction task. My team of a few people used it to label about 15k examples collected from 2TB of input.<p>However, beyond dataset creation, Nosey Parker Explorer was a huge boon for security engineers on engagements. Once you are dealing with more than a few dozen possible findings, it is _hugely_ helpful to be able to interactively slice-and-dice the data. The largest-scale use of it that I&#x27;m aware of was an assumed-breach engagement where we had tens of thousands of potential findings from 20TB of scanned inputs.<p>Nosey Parker Explorer was far too useful to be left as a proof-of-concept or only used internally. A couple weeks ago I released it under the Apache 2 license.<p>If you want to try it, It&#x27;s a Python application (not yet on PyPI). There are prebuilt zipapp releases, but you can also clone the repository and `pip install`. You will first need to use Nosey Parker to scan something. See the project&#x27;s README for details.<p>Happy to answer questions.<p>[1] Nosey Parker: <a href="https:&#x2F;&#x2F;github.com&#x2F;praetorian-inc&#x2F;noseyparker">https:&#x2F;&#x2F;github.com&#x2F;praetorian-inc&#x2F;noseyparker</a><p>[2] Textual: <a href="https:&#x2F;&#x2F;textual.textualize.io" rel="nofollow">https:&#x2F;&#x2F;textual.textualize.io</a><p>[3] DuckDB: <a href="https:&#x2F;&#x2F;duckdb.org" rel="nofollow">https:&#x2F;&#x2F;duckdb.org</a>

no comments

no comments