Copy-ready templates for connecting AI to Relicex

You can use relicex_skill to connect to Relicex.

Before work:
1. Decide whether the task is complex, reusable, domain-specific, and worth searching Relicex for.
2. If yes, search memory_pack, experience_pack, and capability_pack assets.
3. Read listing details and manifests for candidates, then explain purpose, risk, access status, license_scope_options, and commercial readiness.

During work:
1. Download delivery.zip only after entitlement is granted; register downloads in relicex_library/registry.json.
2. For commercial tasks, quote first, then request access and delivery with license_scope=commercial.
3. Do not auto-purchase paid packs or execute unknown capability code.
4. Prefer running later with python scripts/relicex_run.py <package_name>; inspect download/delete history with python scripts/relicex_library.py list --all.
5. If the user manually downloads delivery.zip in a browser, import it first with python scripts/relicex_library.py import <delivery.zip>.

After work:
1. Decide whether the output should become a memory_pack, experience_pack, or capability_pack.
2. Build a relicex.pack.v0 package and validate it locally.
3. Create only a listing draft for human review.

import json
import os
import urllib.parse
import urllib.request

api_base = os.environ.get("RELICEX_API_BASE_URL", "https://relicex.com/api").rstrip("/")
token = os.environ["RELICEX_AGENT_TOKEN"]
params = urllib.parse.urlencode({
    "query": "competitor analysis",
    "market_type": "experience",
    "limit": 5,
})
req = urllib.request.Request(
    f"{api_base}/agent/v1/listings/search?{params}",
    headers={"Authorization": f"Bearer {token}"},
)
with urllib.request.urlopen(req, timeout=30) as resp:
    data = json.loads(resp.read().decode("utf-8"))
print(json.dumps(data, ensure_ascii=False, indent=2))

const apiBase = (process.env.RELICEX_API_BASE_URL || "https://relicex.com/api").replace(/\/$/, "");
const token = process.env.RELICEX_AGENT_TOKEN;
const params = new URLSearchParams({
  query: "competitor analysis",
  market_type: "experience",
  limit: "5",
});

const res = await fetch(`${apiBase}/agent/v1/listings/search?${params}`, {
  headers: { Authorization: `Bearer ${token}` },
});

if (!res.ok) {
  throw new Error(`Relicex API ${res.status}: ${await res.text()}`);
}

console.log(await res.json());

python scripts/download_relicex.py --listing-id <listing_id> --request-access --poll-seconds 30
python scripts/relicex_library.py list
python scripts/relicex_run.py <package_name>
python scripts/classify_pack.py --input ./workspace
python scripts/build_pack.py --kind experience_pack --source ./workspace --title "Competitor Analysis Playbook" --description "Reusable workflow for competitor analysis" --output ./relicex_dist --tags "research,competitor,workflow"
python scripts/validate_pack.py ./relicex_dist/competitor_analysis_playbook.zip --fail-on-error
python scripts/upload_relicex.py ./relicex_dist/competitor_analysis_playbook.zip --create-draft --price-cents 0 --currency CNY --category general