Numbers-LE: Zero Hassle Number Extraction
Pull every number out of the current file in one keystroke
JSON/JSONC, YAML, CSV/TSV, TOML, INI, Environment files, and twelve source languages
Useful? A star or rating is how other developers find it —
★ GitHub ·
★ Open VSX ·
★ Marketplace
What it does
Open a file, press Ctrl+Alt+N (Cmd+Alt+N on Mac), and every numeric value in the document lands in a new editor — deduplicate and sort it from there. Works in VS Code and in VS Code–based editors like Cursor and VSCodium (installable from Open VSX).
- Data validation — pull the numbers out of a config or fixture and eyeball ranges at a glance
- Config audits — compare ports, thresholds, and limits across INI/TOML/.env files
- CSV work — extract everything, one column, or several columns into separate documents
Install
| Where |
What you get |
Install |
| VS Code |
The same extraction, in your editor, on a keystroke |
Marketplace |
| Cursor, VSCodium, Windsurf |
The same extension |
Open VSX |
| A terminal or a CI step |
The same run over a whole tree, with exit codes |
cargo install numbers-le · crates.io |
| Any MCP agent, via Node |
extract_numbers over stdio |
npx numbers-le-mcp · npm |
| Zed |
The MCP server as a context server |
add it by hand (no listing yet) |
Use it from an AI agent
The same engine runs as an MCP server, so an agent can call it directly instead of you running a command.
| Editor |
How |
| VS Code 1.101+ |
Nothing to install — the extension registers extract_numbers with agent mode |
| Zed |
No listing yet — add the MCP server by hand |
| Claude Code |
claude mcp add numbers-le -- npx -y numbers-le-mcp |
| Cursor, Windsurf, anything else |
point it at npx numbers-le-mcp |
extract_numbers(content, format?, filename?, dedupe?, maxResults?)
Returns the values in document order, capped at 500 by default with meta.truncated. A format is optional — anything unrecognised is scanned as plain text rather than refused.
The server takes content and returns data — it reads no files and makes no network requests of its own. Published as numbers-le-mcp on npm and as io.github.nolindnaidoo/numbers-le in the MCP registry.
Configuring it by hand — any host with an MCP config file
Most hosts read a JSON config. Add one entry:
{
"mcpServers": {
"numbers-le": {
"command": "npx",
"args": ["-y", "numbers-le-mcp"]
}
}
}
-y skips the install prompt on first run. Pin a version if you would rather not track releases — numbers-le-mcp@2.3.1.
Prefer not to go through npx on every launch? Install it once and point at the binary instead:
npm install -g numbers-le-mcp
{
"mcpServers": {
"numbers-le": { "command": "numbers-le-mcp" }
}
}
It speaks MCP over stdio and needs no environment variables, no API key and no configuration of its own. To check it before wiring it into anything:
echo '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' | npx -y numbers-le-mcp
That prints the tool list and exits — if you see extract_numbers, the server works.
| Format |
What counts as a number |
| JSON |
Number values anywhere in the structure. Quoted numbers ("42") are strings, not numbers, and are never extracted |
| YAML |
Number values, including hex/octal int forms and exponents; multi-document (---) streams extract from every document |
| TOML |
Integer and float values (dates are not numbers) |
| CSV |
Cells whose entire content is numeric — 19.99 counts, 12abc and 1,000 do not. No header inference: every row is data |
| INI |
Values whose entire content is numeric (INI values are inherently strings) |
| Environment |
Values whose entire content is numeric (PORT=3000 yields 3000; VERSION=1.2.3 yields nothing) |
The shared policy (one heuristics module, applied to every format): only finite numbers are extracted — NaN and inf/.inf never appear in results — and a coerced string must be numeric in full: plain decimals with optional sign, decimal point, and exponent.
Python, Rust, Go, Java, Kotlin, C#, C, C++, JavaScript, TypeScript, SQL and shell are read by a numeric-literal extractor that understands hex 0xFF, binary 0b1010, octal 0o755 and legacy 0755, digit separators 1_000_000 and 1'000, and suffixes 123n, 1.5f, 10u32, 100L — and that knows u32 is a type name, not the number 32. A dialect changes an answer: 0755 is 493 in C, C++, Go and Java, and 755 in Rust, Python 3, Kotlin and C#.
Every finding carries the notation it was written in — decimal, hex, binary, octal, scientific, bigint. For anything left over you can pick a format explicitly or fall back to a plain-text scan (no grammar: v1.2.3 reads as 1.2 and 0.3).
The CLI
The same extraction runs from a terminal or a shell pipeline: a Rust CLI
in crate/, sharing one corpus with the extension —
crate/fixtures/ — so the two can never read a
document differently.
numbers-le . # every number in the tree, as JSON
numbers-le --values config/ # just the numbers, one per line
numbers-le mcp # the same extraction over MCP on stdio
# the point of the whole thing:
numbers-le --values --dedupe src/ | sort -n > after.txt
diff before.txt after.txt # what constants moved this release
Somebody has to verify that the rate in the code is the rate in the
specification. A tax percentage, a retention window, a rounding
boundary. In a regulated setting that check is a deliverable, and the
person doing it usually has no checkout and never has the editor open.
Numbers are printed exactly as JavaScript prints them — 1e+21, not
Rust's 1000000000000000000000. This tool's whole output is numbers as
text, so the rendering is the contract, and the corpus pins it.
Exit codes follow grep — 0 numbers found, 1 none found, 2 the question
was malformed.
Commands
| Command |
Description |
Numbers-LE: Extract Numbers (Ctrl+Alt+N / Cmd+Alt+N) |
Extract all numbers from the active document |
Numbers-LE: Deduplicate Numbers |
Remove duplicate numbers from the results |
Numbers-LE: Sort Numbers |
Sort results numerically or by magnitude |
Numbers-LE: Toggle CSV Streaming |
Flip csv.streamingEnabled for large CSV files |
Numbers-LE: Open Settings |
Open Numbers-LE settings |
Numbers-LE: Help & Troubleshooting |
Built-in documentation |
Settings
| Setting |
Default |
Description |
numbers-le.openResultsSideBySide |
true |
Open results beside the current editor |
numbers-le.postProcess.openInNewFile |
true |
Dedupe/Sort write to a new file instead of replacing the editor content |
numbers-le.copyToClipboardEnabled |
false |
Also copy results to the clipboard (non-CSV) |
numbers-le.dedupeEnabled |
false |
Deduplicate automatically during extraction |
numbers-le.sortEnabled |
false |
Sort automatically during extraction |
numbers-le.sortMode |
off |
numeric-asc/desc, magnitude-asc/desc |
numbers-le.csv.streamingEnabled |
false |
Incremental CSV parsing for large files |
numbers-le.showParseErrors |
false |
Surface parse errors as notifications |
numbers-le.notificationsLevel |
silent |
all = every notification, important = warnings + errors, silent = errors only |
numbers-le.safety.enabled |
true |
Guardrails for very large files/outputs |
numbers-le.safety.fileSizeWarnBytes |
1000000 |
Warn above this file size |
numbers-le.safety.largeOutputLinesThreshold |
50000 |
Offer Open/Copy/Cancel above this result count |
numbers-le.safety.manyDocumentsThreshold |
8 |
Confirm before opening this many column documents |
numbers-le.statusBar.enabled |
true |
Show the status bar item |
numbers-le.telemetryEnabled |
false |
Local-only event log (see Privacy) |
Languages
Twelve languages besides English:
German · Spanish · French · Indonesian · Italian · Japanese · Korean ·
Portuguese (Brazil) · Russian · Ukrainian · Vietnamese · Chinese (Simplified)
Both halves are covered — the manifest (command titles, setting names and
descriptions) and everything shown while the extension runs (notifications,
the status bar, quick-picks and prompts). The extension follows VS Code's
display language, so it matches whatever the editor is already set to; no
setting of its own.
Privacy & security
- No network access. The extension never sends data anywhere. The
telemetryEnabled setting only writes events to a local Output Channel you can inspect (Numbers-LE).
- The MCP server holds the same line. It takes content as an argument and returns data: no filesystem access, no network calls, no telemetry. Your agent already has file-read tools, so duplicating them inside the server would add a path-traversal surface for no capability.
check:mcp-bundle fails the build if the server ever imports something that could reach either.
- Error notifications redact home directories and credential-shaped fragments.
Documentation
| Input |
Size |
Found |
Time |
Rate |
Scan speed |
| JSON metrics |
2.18 MB |
120,000 |
22.32 ms |
5,376,475/sec |
97.9 MB/s |
| CSV data |
0.93 MB |
150,000 |
51.26 ms |
2,926,066/sec |
18.2 MB/s |
| INI config |
0.42 MB |
30,000 |
20.05 ms |
1,496,430/sec |
20.7 MB/s |
Median of 7 runs after warmup, on Apple M5 Pro, 24 GB RAM, Node 24.3.0. Inputs are generated
by scripts/benchmark.ts rather than checked in, so the sizes above are
exactly what was measured. Reproduce with bun run benchmark.
These are machine-specific and are not asserted in CI — a benchmark that gates
a build only tells you how busy the runner was.
Testing
| Metric |
Coverage |
| Statements |
90.65% |
| Branches |
84.08% |
| Functions |
96.34% |
| Lines |
92.42% |
358 test cases across 23 files, plus an integration suite that runs
in a real VS Code extension host and an end-to-end test that installs the
built .vsix into a clean profile.
Generated from a real run — coverage/coverage-summary.json and
coverage/test-results.json — by scripts/coverage-readme.js; CI fails if
this section drifts. Reproduce with bun run test:coverage, and the case
count is the one vitest prints.
More from the LE family
Sixteen single-purpose tools for the work in front of every model. Each ships
a Rust CLI and an MCP server. One page: letools.dev
Get it out
- String-LE — Extract every string in a codebase, with its position, so a person can read them
- Numbers-LE — Extract every hardcoded number in a codebase, so a person can check them
- Units-LE — Extract every quantity with its unit, normalized, and refuse the ambiguous ones by name
- Dates-LE — Extract every date and timestamp, and the exact instant each one resolves to
- IDs-LE — Extract every UUID, ULID, NanoID, ObjectId and Snowflake, and decode the time inside
- IPs-LE — Extract every IP address, CIDR block and MAC, normalized and classified by scope
- URLs-LE — Extract every URL in a codebase, with its protocol and exact position
- Paths-LE — Extract every file path in a codebase, and say whether it still points at anything
- Colors-LE — Extract every color in a codebase, and say which ones are not in your palette
Check it
- Regex-LE — Find every regex in a codebase, and report which can be driven into catastrophic backtracking
- Versions-LE — Find where one dependency is constrained differently across a repository's manifests
- i18n-LE — Identify the i18n library a project uses, then audit its catalogs by that library's rules
- Scrape-LE — Check whether a page is scrapeable before the scraper is written, and say when it cannot tell
Guard it
- Secrets-LE — Find hardcoded credentials in a codebase, and never print one into the report
- EnvSync-LE — Compare the dotenv files in a tree, and say which keys are missing from which
- Unicode-LE — Find the Unicode that hides meaning — bidi controls, invisibles, homoglyphs, mixed scripts
Each stands on its own: no shared crate, no published core. Where two of them
agree, it is because the same answer was right twice.
Contact — nolindnaidoo.com · GitHub · LinkedIn
Also by nolindnaidoo
Rust — pixelcoords and pixelactions are one loop: pixelcoords answers
where, pixelactions acts there. Their own tools, their own voice — not
part of the LE family.
License
MIT © nolindnaidoo