lokalise-common-errors

What this skill does

# Lokalise Common Errors

## Overview

Every Lokalise API error returns a JSON body with a consistent structure. This skill covers the error response format, diagnosis of each HTTP status code (401, 400, 404, 429, 413, 500, 503), diagnostic curl commands for rapid troubleshooting, and a reusable error handling wrapper for the Node SDK.

## Prerequisites

- `curl` available for diagnostic commands
- `@lokalise/node-api` SDK installed for the error wrapper
- API token stored in `LOKALISE_API_TOKEN` environment variable
- `jq` installed for parsing JSON responses (optional but recommended)

## Instructions

### 1. Understand the Error Response Format

All Lokalise API errors return this structure:

```json
{
  "error": {
    "message": "Human-readable error description",
    "code": 401
  }
}
```

The `code` field mirrors the HTTP status code. The `message` field provides specifics. When using the Node SDK, errors are thrown as exceptions with `error.code`, `error.message`, and `error.headers` properties.

### 2. Diagnose by Status Code

#### 401 Unauthorized — Invalid or Missing API Token

```json
{"error": {"message": "Invalid `X-Api-Token` header", "code": 401}}
```

**Causes:**

- Token is incorrect, expired, or revoked
- `X-Api-Token` header missing from request
- Token copied with leading/trailing whitespace

**Fix:**

```bash
# Verify your token works
curl -s -o /dev/null -w "%{http_code}" \
  -H "X-Api-Token: ${LOKALISE_API_TOKEN}" \
  "https://api.lokalise.com/api2/teams"

# Expected: 200
# If 401: regenerate token at https://app.lokalise.com/profile#apitokens
```

```bash
# Check for whitespace in token
echo -n "$LOKALISE_API_TOKEN" | xxd | head -2
# Look for 0a (newline) or 20 (space) at start/end
```

#### 400 Bad Request — Validation Errors

```json
{"error": {"message": "Invalid parameter `platform` - must be one of: ios, android, web, other", "code": 400}}
```

**Common 400 causes:**

- Invalid `project_id` format (must be `{number}.{alphanumeric}`)
- Missing required fields in POST/PUT body
- Invalid language ISO code
- Key name exceeding 256 characters
- Invalid platform value (must be `ios`, `android`, `web`, or `other`)

**Fix:**

```bash
# Validate project ID format
curl -s -H "X-Api-Token: ${LOKALISE_API_TOKEN}" \
  "https://api.lokalise.com/api2/projects/${PROJECT_ID}" | jq '.project_id'

# List valid languages for a project
curl -s -H "X-Api-Token: ${LOKALISE_API_TOKEN}" \
  "https://api.lokalise.com/api2/projects/${PROJECT_ID}/languages" \
  | jq '.languages[].lang_iso'
```

#### 404 Not Found — Resource Does Not Exist

```json
{"error": {"message": "Project not found", "code": 404}}
```

**Causes:**

- Project ID is wrong or project was deleted
- Key, translation, or webhook ID does not exist
- Token lacks access to the specified project (appears as 404, not 403)

**Fix:**

```bash
# List all accessible projects to find the correct ID
curl -s -H "X-Api-Token: ${LOKALISE_API_TOKEN}" \
  "https://api.lokalise.com/api2/projects?limit=100" \
  | jq '.projects[] | {project_id, name}'

# Verify a specific key exists
curl -s -H "X-Api-Token: ${LOKALISE_API_TOKEN}" \
  "https://api.lokalise.com/api2/projects/${PROJECT_ID}/keys/${KEY_ID}" \
  | jq '.key_id // .error'
```

#### 429 Too Many Requests — Rate Limited

```json
{"error": {"message": "Too many requests", "code": 429}}
```

The response includes a `Retry-After` header indicating seconds to wait.

**Fix:** See `lokalise-rate-limits` skill for full implementation. Quick recovery:

```bash
# Check current rate limit status on any request
curl -s -D - -o /dev/null \
  -H "X-Api-Token: ${LOKALISE_API_TOKEN}" \
  "https://api.lokalise.com/api2/projects" 2>&1 \
  | grep -i "x-ratelimit\|retry-after"

# Output:
# X-RateLimit-Limit: 6
# X-RateLimit-Remaining: 5
# X-RateLimit-Reset: 1700000001
```

#### 413 Payload Too Large — Request Body Exceeds Limit

```json
{"error": {"message": "Request entity too large", "code": 413}}
```

**Causes:**

- File upload exceeds 50 MB
- Bulk key creation with too many keys in one request (max 500)
- Translation value exceeds character limit

**Fix:**

- Split bulk operations into batches of 500 items
- Compress files before upload or split into smaller files
- For large translation values, check if the content truly belongs in a translation key

#### 500 Internal Server Error / 503 Service Unavailable

```json
{"error": {"message": "Internal server error", "code": 500}}
```

**These are Lokalise-side issues.** Do not retry immediately in a tight loop.

**Fix:**

```bash
# Check Lokalise status page
curl -s "https://status.lokalise.com/api/v2/status.json" | jq '.status'

# If status is operational, retry after 30 seconds
# If status shows incident, wait for resolution
```

Retry strategy for 500/503: wait 30 seconds, retry up to 3 times, then alert.

### 3. Run Diagnostic Commands

Quick health check script to diagnose the most common issues in sequence:

```bash
#!/bin/bash
# lokalise-diagnose.sh — Run against your environment
TOKEN="${LOKALISE_API_TOKEN}"
PROJECT="${LOKALISE_PROJECT_ID}"

echo "=== 1. Token validation ==="
STATUS=$(curl -s -o /dev/null -w "%{http_code}" \
  -H "X-Api-Token: $TOKEN" \
  "https://api.lokalise.com/api2/teams")
if [ "$STATUS" = "200" ]; then
  echo "Token: VALID"
else
  echo "Token: INVALID (HTTP $STATUS)"
  exit 1
fi

echo "=== 2. Project access ==="
curl -s -H "X-Api-Token: $TOKEN" \
  "https://api.lokalise.com/api2/projects/$PROJECT" \
  | jq '{project_id: .project_id, name: .name, team_id: .team_id}'

echo "=== 3. Rate limit status ==="
curl -s -D /dev/stderr -o /dev/null \
  -H "X-Api-Token: $TOKEN" \
  "https://api.lokalise.com/api2/projects" 2>&1 \
  | grep -i "x-ratelimit"

echo "=== 4. Key count ==="
curl -s -H "X-Api-Token: $TOKEN" \
  "https://api.lokalise.com/api2/projects/$PROJECT/keys?limit=1" \
  | jq '.project_id as $p | {project: $p, total_keys: .keys | length}'
```

### 4. Build a Reusable Error Handling Wrapper

Wrap all SDK calls with structured error handling:

```typescript
import { LokaliseApi } from "@lokalise/node-api";

interface LokaliseError {
  code: number;
  message: string;
  headers?: Record<string, string>;
}

function isLokaliseError(error: unknown): error is LokaliseError {
  return (
    typeof error === "object" &&
    error !== null &&
    "code" in error &&
    "message" in error
  );
}

async function lokaliseCall<T>(
  fn: () => Promise<T>,
  context: string
): Promise<T> {
  try {
    return await fn();
  } catch (error) {
    if (!isLokaliseError(error)) throw error;

    switch (error.code) {
      case 401:
        throw new Error(
          `[${context}] Authentication failed. ` +
          `Regenerate token at https://app.lokalise.com/profile#apitokens`
        );
      case 400:
        throw new Error(
          `[${context}] Invalid request: ${error.message}. ` +
          `Check parameter values and required fields.`
        );
      case 404:
        throw new Error(
          `[${context}] Resource not found: ${error.message}. ` +
          `Verify the project/key/resource ID exists and token has access.`
        );
      case 429:
        console.warn(`[${context}] Rate limited. See lokalise-rate-limits.`);
        throw error; // Let the rate limit handler deal with retries
      case 413:
        throw new Error(
          `[${context}] Payload too large: ${error.message}. ` +
          `Split into smaller batches (max 500 items per request).`
        );
      case 500:
      case 503:
        throw new Error(
          `[${context}] Lokalise server error (${error.code}). ` +
          `Check https://status.lokalise.com — retry after 30s.`
        );
      default:
        throw new Error(
          `[${context}] Lokalise error ${error.code}: ${error.message}`
        );
    }
  }
}

// Usage
const lokalise = new LokaliseApi({ apiKey: process.env.LOKALISE_API_TOKEN! });

const keys = await lokaliseCall(
  () => lokalise.keys().list({ project_id: projectId, limit: 5