Developer reference

API

Programmatic access to the prompt library. Query classifications, fetch prompt content, and build on the same data the platform renders.

You're helping me explore MLAD's pre-classified prompt corpus at
https://mlad.ai. Fetch https://mlad.ai/prompts/llms.txt first for the
five-axis taxonomy, then https://mlad.ai/api for the HTTP contract.

Ground rules:
- Base URL: https://mlad.ai/api/v1/prompts (list; /{id} for detail).
- Auth: Authorization: Bearer <my key> -- I'll paste this next.
- User-Agent: set to something descriptive (e.g. "mlad-explorer/0.1").
  Default library UAs (like Python-urllib) get a 403 before auth.
- Filters repeat to OR within an axis; axes AND across. limit<=100,
  offset<=10000. On 401 stop. On 429 back off; honor Retry-After.
- If a fetch fails with a sandbox host_not_allowed / allowlist error,
  tell me to add mlad.ai to Settings -> Capabilities -> Code execution
  -> Additional allowed domains and open a fresh chat.

When I give you the key, ask me what I want to search for. Propose a
filter shape, call the API, and summarise what you find with links
to the canonical /prompts/<source>/<slug> pages for anything useful.
curl https://mlad.ai/api/v1/prompts \
  -H "Authorization: Bearer mlad_live_XXXXXXXXXXXX_YYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYY"
# Prompts tagged with the "debugging" theme (3,750 matches as of today)
curl "https://mlad.ai/api/v1/prompts?themes=debugging" \
  -H "Authorization: Bearer mlad_live_…"
# Python OR TypeScript, AND tagged "testing"
curl "https://mlad.ai/api/v1/prompts?techStack=python&techStack=typescript&themes=testing" \
  -H "Authorization: Bearer mlad_live_…"
# All System prompts that are Persistent + Global (agent-identity patterns)
curl "https://mlad.ai/api/v1/prompts?promptTypes=System&activations=Persistent&scopes=Global" \
  -H "Authorization: Bearer mlad_live_…"
curl "https://mlad.ai/api/v1/prompts?searchText=refactor&limit=10" \
  -H "Authorization: Bearer mlad_live_…"
curl "https://mlad.ai/api/v1/prompts?themes=debugging&limit=50&offset=100" \
  -H "Authorization: Bearer mlad_live_…"
# Use an id from the list response
curl https://mlad.ai/api/v1/prompts/prompt-4e51e0f6bf5b \
  -H "Authorization: Bearer mlad_live_…"
import json
import urllib.request

req = urllib.request.Request(
    "https://mlad.ai/api/v1/prompts?limit=5",
    headers={
        "Authorization": "Bearer mlad_live_XXXXXXXXXXXX_YYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYY",
        "User-Agent": "my-app/0.1 (+https://example.com)",
    },
)
with urllib.request.urlopen(req) as resp:
    data = json.load(resp)

print(data["total"], "matches; first:", data["prompts"][0]["title"])

Sign in to get started

Free to sign up. API access requires a paid plan.

Sign in to get started