GigTry API: Automate the 89 Tools from Your Own Systems
The GigTry API lets external automated systems run the tools programmatically: create an API key in your dashboard, send the key as a Bearer token to POST /api/v1/execute with a tool slug and input JSON, and receive the same output the web studio produces. Keys carry scopes, per-key rate limits and usage tracking, and the GET reference is self-documenting.
How to make your first API call
Create an API key
In your account dashboard open API Keys, create a key with the tools:execute scope, and copy it immediately — the full key is shown once, only its prefix is stored.
Call the execute endpoint
POST to /api/v1/execute with the Authorization: Bearer header, a JSON body carrying the tool slug and its input object. The response contains the generated output plus provider metadata.
Handle the responses
200 returns output; 400 means invalid input and includes the expected shape and example; 401/403 are auth or scope problems; 429 is rate-limited with a Retry-After header; 503 means all providers were busy — retry with backoff.
curl -X POST https://gigtry.com/api/v1/execute \
-H "Authorization: Bearer gt_live_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"tool": "social-media-post-generator",
"input": {
"topic": "launch of our eco-friendly water bottle",
"platform": "linkedin",
"tone": "professional"
}
}'Generate a LinkedIn post from your automation system
What can the API do besides execution?
GET /api/v1/tools lists the full catalog with executable flags and credit costs — your system can discover tools dynamically instead of hardcoding slugs. GET /api/v1/usage reports per-key consumption for billing and quota dashboards. GET /api/v1/execute without a body returns the complete self-documenting reference: every curated tool with its input schema, expected fields and a ready-made example payload.
| Endpoint | Method | Scope | Purpose |
|---|---|---|---|
| /api/v1/execute | POST | tools:execute | Run any executable tool |
| /api/v1/tools | GET | tools:read | Catalog with schemas and costs |
| /api/v1/usage | GET | usage:read | Per-key usage and quota |
| /api/api-keys | POST | dashboard session | Create keys (max 10 per account) |
Machine-readable catalog for agents
AI agents should read /llms.txt and /llms-full.txt first — they contain the tool inventory, the execution contract and high-intent question-to-tool mappings in plain text, which is the fastest way for an autonomous system to plan multi-tool workflows.
Frequently asked questions
How do I authenticate API requests?
Create a key in the dashboard, then send it as an Authorization: Bearer header. Keys are stored as SHA-256 hashes with only a visible prefix, support soft revocation, and can carry the scopes tools:read, tools:execute and usage:read.
What are the API rate limits?
Each key carries its own configurable rate limit, enforced server-side with a concurrency guard that returns 503 and a Retry-After header when capacity is saturated. Usage is metered per key and visible at /api/v1/usage.
Which tools can the API execute?
Every text-friendly tool — 61+ of the 89 at launch, including nine curated executors with tuned prompts and validated input schemas. The catalog endpoint marks each tool with an executable flag so your integration can adapt automatically.
Does API usage cost credits?
Yes — API executions consume the same credit balance as the web studio, at the same per-tool rates. There is no separate API fee; enterprise-scale volume simply maps to the Enterprise plan or credit packs.
Can API keys be scoped and revoked?
Keys carry granular scopes, optional expiry dates, and soft revocation — revoked keys stop working instantly while preserving the audit trail of past usage. All key management happens in the dashboard without contacting support.
Was this guide what you needed?