Introducing Agents.·Read
OverviewAgentsWorkflowsExecutionDeveloper ToolsSecurityLibrary
Updated Oct 16, 2026

Utilities

Search source files with ripgrep

ripgrep-search is a source-search node that returns native ripgrep regex events and offsets as JSONL.

Agent
Locate a recurring identifierCollect source locationsHandle no matches explicitly

overview

What ripgrep-search does

Search a copied folder with a case-sensitive Rust regular expression. Native JSONL events identify matching paths, line content and byte offsets, followed by summary statistics. Hidden text is included; repository ignore and config files are not loaded. Generated .git, node_modules, .venv and target folders are excluded.

Native Rust regex JSONL, hidden text included, config/ignore files disabled; 1000 files, 1 MiB/file, 20 MiB, 100 matching lines/file, 60 seconds

No matches is a valid native summary. Invalid regexes and runtime/input failures fail the operation; the source remains unchanged.

This operation uses 15.2.0. Upstream license: MIT. The source link identifies the program used by this node.

source github.com/BurntSushi/ripgrep/tree/15.2.0

use cases

Where ripgrep-search fits

Locate a recurring identifier

Find every matching line with a case-sensitive Rust regex.

Collect source locations

Use native paths and byte offsets to point a later review at the matching text.

Handle no matches explicitly

Treat a zero-match summary as a result instead of a task failure.

reference

ripgrep-search inputs and flags

2 inputs
NameTypeFlagDescription
sourceFOLDER--sourceFolder, at most 1000 files, 1 MiB per file and 20 MiB total. Symlinks refused. Generated .git/node_modules/.venv/target folders excluded; hidden text included; binary files use native handling.
patternSTRING--patternRust regex, 1-4096 UTF-8 bytes. Case sensitive. At most 100 matching lines per file; 60 seconds and 16 MiB output. No matches is a valid native summary.

Showing key inputs. ripgrep-search exposes 2 inputs in total.

example

Run ripgrep-search

ripgrep-search · command
python3 /app/run.py --source ./source --pattern 'needle' ./output
sample output
{
  "type": "summary",
  "data": {
    "stats": {
      "matched_lines": 4,
      "matches": 5
    }
  }
}

guidance

Choosing ripgrep-search

Use ripgrep-search when later steps need structured regex matches and offsets. Fixed-string filters are simpler for literal line selection; this operation does not parse source syntax.

fgrep-by-string

Selects literal matching lines rather than native regex JSONL events.

faq

ripgrep-search questions

source: Folder, at most 1000 files, 1 MiB per file and 20 MiB total. Symlinks refused. Generated .git/node_modules/.venv/target folders excluded; hidden text included; binary files use native handling. pattern: Rust regex, 1-4096 UTF-8 bytes. Case sensitive. At most 100 matching lines per file; 60 seconds and 16 MiB output. No matches is a valid native summary.

JSONL search events. No matches is a valid native summary. Invalid regexes and runtime/input failures fail the operation; the source remains unchanged.

Native Rust regex JSONL, hidden text included, config/ignore files disabled; 1000 files, 1 MiB/file, 20 MiB, 100 matching lines/file, 60 seconds

Use ripgrep-search when later steps need structured regex matches and offsets. Fixed-string filters are simpler for literal line selection; this operation does not parse source syntax.

The image repository also contains a digest-bound source/notices artifact. Download it with oras pull docker.io/nenadzaric/skills-intel-ripgrep-search@sha256:35f4df48f3c3c4f437735ead72b7482b8ce37c61ce63366da61494809cc68d86. The bundle includes the build recipe, licenses, copied notices and component-source directions. Bundle SHA256: f7a211a88db24a2a8f70432fd0229c70c49b068898ea0f6d2d3cfcd24e8bfe3d. Follow its source directions for upstream archives and patches. Image and source/notices repository.

related

More Utilities tools

  1. browser-fetch

    Headless Chromium render: requested URL, final URL, status, and title next to the page.

  2. cewl

    Spider a URL and return a wordlist for password crackers.

  3. csvkit

    SQL on one CSV. The table name is the file stem.

  4. dnsgen

    Wordlist and mined-word subdomain permutation.

  5. duckdb

    In-process SQL on CSV, JSON, and Parquet. No server.

  6. edge-tts

    Microsoft Edge neural speech from text or a script file. No API key.

Run ripgrep-search yourself

The synthetic run checks jsonl search events. The excerpt shows selected native fields asserted by saved passing platform verifier; reconstructed excerpt, not full raw artifact. The command shown is the container entrypoint with example paths, not a host-installed CLI command.

Open ripgrep-search in the libraryBuild a workflow with it

Facts on this page come from the live Trickest tool library.