Question 1

How do I fix broken or malformed JSON online?

Accepted Answer

yellorn ships a 'Data Doctor' best-effort parser with 17 documented auto-fix passes for JSON: it strips BOMs and markdown code fences, repairs smart quotes, removes // and /* */ comments, normalizes Python literals (None/True/False), converts single quotes and unquoted keys, drops trailing commas, repairs missing commas, auto-closes truncated brackets, and more. Its safe Python repr bridge also handles pprint mappings and sets, raw/triple/bytes strings, scalar constructors, and dataclass-like values without eval. Every fix is labelled in the status bar so you consent to every change before publishing.

Question 2

Can I clean up JSON returned by ChatGPT, Claude, or other LLMs?

Accepted Answer

Yes — paste the LLM response directly. yellorn auto-strips markdown ```json fences, removes inline comments, normalizes Python-style booleans (True → true, None → null), and recovers truncated payloads. The status bar shows exactly which fixes ran (e.g. 'Auto-fixed: markdown-code-blocks, trailing-commas, single-quotes') so the transformation is auditable.

Question 3

Does it repair broken XML, YAML, and CSV too?

Accepted Answer

Yes. The same recovery ladder runs across all four formats: 5 fixes for XML (markdown fences, BOM, missing brackets, auto-close-at-EOF, unescaped ampersands), 6 for YAML (BOM, fences, tabs-to-spaces, smart quotes, missing colon before block, unbalanced flow brackets), and 3 for CSV (BOM stripping for Excel-on-Windows exports, fences, smart quotes). Plus the 23-pass JSON ladder, that's 37 documented fix classes total.

Question 4

What file formats can I open and view in yellorn?

Accepted Answer

yellorn handles 25 formats in one editor. Text & structured formats include JSON, JSON Lines (NDJSON), XML, YAML, TOML, CSV, ASCII Table, SQL, Python, cURL commands, Cookie headers, and Raw Text. Canvas TSX files open as source-preserving React previews, not structured-data conversions. Binary data files decode locally in a worker: Apache Parquet, Apache Arrow / Feather, Excel, MessagePack, CBOR, Protobuf, BSON, and gzip. AI / ML artifacts include Safetensors and GGUF model headers, and ComfyUI workflows extracted from PNG. The X.509 certificate / key family includes PEM, DER, PKCS#12 / PFX, PKCS#7, and CSR files with expiry, algorithm, fingerprint, and warning summaries. Drop a file onto the editor; its extension and contents determine the format.

Question 5

How can I preview a .canvas.tsx file online or in VS Code?

Accepted Answer

Drop a .canvas.tsx file onto the yellorn editor or new-tab button, or select Canvas TSX for pasted source. In the Yellorn VS Code extension, run Yellorn: Preview Canvas or use the file's preview button. The file must default-export a React component. A local sandboxed preview supports core cursor/canvas layouts, tables, controls, and charts, plus React imports. Compatibility is partial: other packages, IDE actions, and persistent Canvas state are unavailable. Only preview files you trust; the sandbox is not a CPU or memory limit.

Question 6

Can I open a Parquet, Excel, or Arrow file online without uploading it to a server?

Accepted Answer

Yes. Drop a .parquet, .xlsx, .arrow / .feather, .msgpack, .cbor, .pb, or .bson file onto the editor, or choose the format and select Open file. yellorn decodes it locally in a web worker, then shows a JSON preview you can browse as a tree, graph, or table. Parquet and single-sheet Excel open straight into the sortable table view; multi-sheet workbooks open as a tree keyed by sheet name. Large files are previewed with a byte cap so the tab stays responsive — nothing is uploaded to a server to view it.

Question 7

Can I inspect Safetensors, GGUF, or ComfyUI files?

Accepted Answer

Yes. Open a .safetensors file and yellorn reads just the JSON header — tensor names, dtypes, shapes, and a computed parameter-count summary — never the multi-GB weights. Open a .gguf llama.cpp model and it surfaces the typed metadata KV map, per-tensor shape/type list, and a summary. Drop a ComfyUI-generated PNG and it extracts the embedded node graph (the API prompt + UI workflow) plus a recipe summary with the checkpoint, LoRAs, sampler settings, image dimensions, and resolved positive / negative prompts.

Question 8

Can I decode a PEM, CRT, PFX, or CSR certificate file and check its expiry online?

Accepted Answer

Yes. Drop a .pem, .crt, .cer, .der, .pfx / .p12, .p7b, or .csr file onto the editor, or choose Certificate / Key and select Open file. yellorn decodes it locally in a web worker — nothing is uploaded. Every certificate type produces the same at-a-glance '__insight__' summary: what it is, the subject and issuer, whether it's expired or expiring within 30 days, the public-key and signature algorithm, the SHA-256 fingerprint, and warnings (self-signed, weak RSA key, SHA-1 signature, password-protected PKCS#12). Below the summary you get the full decoded detail — Subject Alternative Names, key usage, extended key usage, basic constraints, and validity window. If a PKCS#12 / PFX bundle is encrypted, yellorn prompts you for its password and unlocks the certificate details right in the browser — the password and the file never leave your device. Private-key material is never decoded or displayed — only the key's algorithm and format.

Question 9

Does yellorn format and beautify SQL queries (MySQL, Postgres, Athena, BigQuery, Snowflake)?

Accepted Answer

Yes. Pick the SQL format from the toolbar, paste your query, and hit Smart Format (Cmd+Shift+F) to auto-indent and uppercase keywords with sql-formatter — supports MySQL, PostgreSQL, MS SQL Server, Amazon Athena (Trino/Presto), Google BigQuery, Snowflake, MariaDB, IBM DB2, and SQLite dialects. Simple Format preserves your existing keyword case, Safe Format normalises whitespace only, and Compress minifies to a single line for embedding in code. Best-effort repair strips markdown ```sql fences, BOMs, smart quotes, and zero-width characters from rich-text pastes automatically.

Question 10

How do I convert JSON to XML, YAML, or CSV (and back)?

Accepted Answer

Open the Format selector in the editor toolbar and choose a target row marked Convert. yellorn parses your input to a canonical value and rewrites the tab in the selected format. If the target is stricter — such as CSV with nested cells or XML with a top-level list — the row is marked Lossy and a 'Convert anyway' warning explains the exact caveat before anything changes. A single undo restores both the original text and format, and Python is available as a first-class conversion target.

Question 11

Can I visualize JSON, XML, YAML, or CSV as an interactive tree, graph, or table?

Accepted Answer

Yes — structured data has three view modes: a virtualized tree (default for JSON/XML/YAML), a node-link graph powered by React Flow + Dagre layout, and a sortable table view (default for CSV). Canvas TSX has its own React preview rather than these data views. Each format declares which views it supports; the visualizer toolbar filters automatically. Cmd+Shift+T cycles through the available views.

Question 12

Is there a JMESPath playground or query tool?

Accepted Answer

Yes. Type a JMESPath expression in the visualizer search bar (e.g. users[?age > `30`].name). yellorn evaluates the expression, swaps the effective tree to the matching nodes, shows match counts, and provides Prev/Next navigation. JMESPath ships as a 20 KiB lazy chunk that only loads when you first open the search bar.

Question 13

How can I publish a payload as a mock REST API or webhook?

Accepted Answer

Sign in (free), click Publish in the editor toolbar, and yellorn returns a globally distributed HTTPS URL. Anyone you share it with can GET / POST / PUT / PATCH / DELETE against the slug; you watch the request log fill in real time from the owner-only portal at /webhook/<slug>. Configure the canned response status code (200/401/429/500) and custom response headers to simulate failure paths without redeploying anything. Paid tiers can choose a custom, memorable endpoint path (e.g. /api/webhook/my-stripe-hook) instead of a random slug. Free tier: 5 active URLs, each live for 24 hours. Pro: 1,000 new URLs/month, active while your subscription is active.

Question 14

Can my mock API return dynamic data — a fresh UUID, timestamp, or random value on every request?

Accepted Answer

Yes — yellorn's dynamic tokens let a published webhook response (and Request Sender bodies, URLs, and headers) compute part of its value at the moment the request is served. Wrap a small Python-flavoured expression in double curly braces and it renders live on every hit: {{ uuid4() }} for a fresh RFC-4122 UUID, {{ now() }} or {{ now() + timedelta(days=1) }} for ISO timestamps, {{ randint(1, 100) }} / {{ randfloat(0, 1) }} / {{ choice("active", "pending") }} for random values, {{ objectid() }} and {{ snowflake() }} for ids, and webhook responses can echo the caller with {{ request.method }}, {{ request.ip }}, {{ request.header("User-Agent") }}, or {{ request.query("id") }}. The webhook portal and Request Sender include an Insert-token picker with live previews, in-editor autocomplete, inline highlighting, and a Preview toggle. Every template is validated before it's saved, so a malformed token never ships.

Question 15

Can I compare two JSON files for differences?

Accepted Answer

Yes — yellorn's Compare Mode lets you select two tabs from the tab bar and run them through Monaco's read-only diff editor. JSON content is sorted before diffing for stable, semantically-meaningful output (key reorderings don't pollute the diff). Useful for API response regression, fixture consistency checks, and customer-payload triage.

Question 16

Can I send HTTP requests from yellorn like Postman?

Accepted Answer

Yes — yellorn's Request Sender lets you compose and dispatch GET, POST, PUT, PATCH, DELETE, HEAD, and OPTIONS requests directly from your browser. Configure headers, auth (Bearer, Basic, custom), request body (JSON, text, form-urlencoded), and save templates for reuse. The live result includes response headers and up to 1 MB of body data; recent history keeps status, timing, and a short body preview. SSRF protection and a loop guard prevent abuse.

Question 17

Can I sync my workspace across devices?

Accepted Answer

Yes — Cloud Backup lets signed-in users back up open editor tabs to the cloud, then restore them on any device. Palette, pane layout, and other UI preferences stay local. Open Settings and use Back up to Cloud / Restore from Cloud.

Fix, Convert & View 25 Data Formats — JSON, CSV, Parquet, Excel & Mock Webhook APIs · yellorn

Universal data-file viewer — 25 formats, from JSON to Parquet to Safetensors to X.509 certificates

Data Doctor — repair broken structured data instantly

SQL auto-format — MySQL, Postgres, Athena, BigQuery, Snowflake

Cross-format converter — JSON ↔ XML ↔ YAML ↔ CSV

Mock webhook REST API — publish payloads as live endpoints

Dynamic mock responses — live UUIDs, timestamps & random data with `{{ }}` tokens

Multi-view visualizer + JMESPath search

Request Sender — compose and dispatch HTTP requests

Cloud Backup — editor tabs across devices

Compare mode + smart format ladder

Waiting for data

Waiting for data

Fix, Convert & View 25 Data Formats — JSON, CSV, Parquet, Excel & Mock Webhook APIs · yellorn

Universal data-file viewer — 25 formats, from JSON to Parquet to Safetensors to X.509 certificates

Data Doctor — repair broken structured data instantly

SQL auto-format — MySQL, Postgres, Athena, BigQuery, Snowflake

Cross-format converter — JSON ↔ XML ↔ YAML ↔ CSV

Mock webhook REST API — publish payloads as live endpoints

Dynamic mock responses — live UUIDs, timestamps & random data with {{ }} tokens

Multi-view visualizer + JMESPath search

Request Sender — compose and dispatch HTTP requests

Cloud Backup — editor tabs across devices

Compare mode + smart format ladder

Yellorn data editor workspace

Waiting for data

Waiting for data

Dynamic mock responses — live UUIDs, timestamps & random data with `{{ }}` tokens