Error Memory

Error Memory is where a team keeps the errors it has hit and how it fixed them. Paste an error message or a whole stack trace: it recognises the language and the error, masks what changes between occurrences (line numbers, IDs, timestamps, paths, values) and finds the same error, or similar ones, in your team’s library, with the fixes that worked.

Features

Recognises JavaScript and TypeScript (V8 and Firefox stacks), Python tracebacks, Java and Kotlin (root cause from “Caused by”), Go panics, .NET, Ruby, Rust panics, PHP, PostgreSQL, MySQL and SQLite errors, HTTP problem+json and JSON error bodies
Fingerprints that ignore line numbers, IDs, UUIDs, timestamps, IPs, URLs, file paths and quoted values, so the same error matches across machines and releases
Similarity search for near misses, plus word search across titles, messages, services, tags and fixes
Log an error with its fix or workaround (Markdown), links to the PR or ticket and the version it was fixed in; duplicates are spotted before you save
Secrets and personal data (tokens, keys, emails, cards, IPs) are removed from errors as you log them
“I hit this too” sightings with environment and version, several fixes per error ranked by “worked for me” votes, and a discussion thread
Status per error (not solved, workaround, resolved) and a “came back” warning when it is seen again after its fix
Shared with your team through a team workspace, encrypted in the browser before it syncs; personal and offline without an account

How to use it

Paste the error or stack trace into the search box.
Open the best match to read how it was fixed, and mark the fix that worked for you.
Nothing found? Solve it, then log it with Log an error so the next person finds it.

FAQ

How does it know two errors are the same?

It parses the error (language, type, message, stack frames) and masks the parts that change between occurrences: numbers, IDs, timestamps, paths, URLs and data values in quotes, while keeping names that identify the bug, such as the property, column or key. Errors with the same type and masked message share a fingerprint. Different but related errors are ranked by how much of the message, the type and the application’s own stack frames they share.

Who can see the errors I log?

Without a team, only you: the library is kept in your browser. In a team workspace every member sees it. The library is encrypted in the browser with the team key before it syncs, so our server stores only ciphertext.

What if an error contains passwords or customer data?

When you log an error, tokens, API keys, private keys, emails, card numbers, phone numbers and IP addresses are replaced with [REDACTED] before it is saved (you can turn this off). Check the text before saving: pattern matching cannot catch everything, such as names.

How is this different from Sentry?

Sentry collects errors from running apps and groups them. Error Memory keeps what your team learned about them: the fix, the workaround, who solved it and whether it worked for others. It also covers errors no monitoring sees, from local builds, CI, migrations and third-party APIs.

More developer tools

API Studio

Build, test and automate HTTP requests right in your browser.

API Studio: Build, test and automate HTTP requests: collections, environments, tests and a runner.

Diagrams

Draw cloud architectures and watch the data flow.

Architecture Builder: Draw cloud architectures and data flows with AWS, Azure, GCP and 1,000+ components.

CI/CD

See, edit and check your CI/CD pipelines.

Pipeline Studio: See and edit your Bitbucket pipelines as a metro map, with checks for mistakes Bitbucket would reject.

Data

Query, browse and edit data files and SQLite databases.

Database Studio: Query, browse and edit CSV, Excel, Parquet and SQLite files with SQL, right in your browser. Nothing is uploaded.

System Design

Simulate traffic through a system design and find its limits.

System Design Simulator: Design a system, send it traffic and watch where it melts: latency, errors and the bottleneck.

Team Knowledge

Every error your team has hit, and how it was fixed.

Error Memory: Search every error your team has hit and how it was fixed; log new ones with their fix.

Formatters & Beautifiers

Make messy payloads, queries and markup readable.

JSON & Log Formatter: Repair, beautify and deep-decode JSON, Python dicts and log lines.
SQL Formatter: Pretty-print queries for 12 SQL dialects.
XML / HTML Beautifier: Beautify or minify HTML and XML, with validation.
JSON Diff: Compare two payloads side by side or as a change list.
SQL Diff: Compare queries, scripts or schemas as SQL, with schema changes and the migration.
JQ / JSONPath: Filter JSON with jq or JSONPath expressions.

Encoders & Decoders

Look inside tokens, encodings and escaped text.

JWT Debugger: Decode tokens, check expiry and verify HMAC signatures.
Base64 / Hex / Binary: Convert text and files between Base64, hex and binary.
URL / Query Parser: Split, edit and rebuild URLs and query strings.
HTML & Unicode Entities: Encode and decode HTML entities and Unicode escapes.

Generators

IDs, hashes, keys, placeholder text and typed code.

UUID / ULID / Hash: Generate UUIDs and ULIDs; hash text or files.
Lorem Ipsum: Placeholder paragraphs, sentences or words.
RSA / EC Key Pair: RSA and EC key pairs as PEM or JWK, via WebCrypto.
Schema & Code Generator: Turn sample JSON into TypeScript, Zod, Go, Rust, SQL and more.

Converters

Move data between formats, times and color spaces.

Unix Timestamp ↔ Date: Convert Unix timestamps to dates and back.
Color Converter: HEX, RGB, HSL and OKLCH with WCAG contrast checks.
Cron Parser: Explain cron expressions and list upcoming runs.
Format Converter: Convert between JSON, YAML, TOML, XML and CSV.

API & Debugging

Reproduce requests, read errors and sanitise logs.

Log → cURL: Rebuild a runnable cURL from logs, HAR or fetch calls.
HTTP Status Codes: Status codes with causes and fixes.
Payload Redactor: Strip secrets and personal data before you share.
Regex Tester: Test patterns live and explain them token by token.