Prompts: Fetch Policy

Fetch Policies

Each policy is optimized for a different deployment scenario.

• MATERIALIZED_FIRST (default)

• ALWAYS_FETCH

• CACHE_TTL

• MATERIALIZED_ONLY

What You Can Do Now

• Configure different fetch policies per environment (development, staging, production) while keeping the same prompt IDs in code.

• Run fully offline in sensitive environments using MATERIALIZED_ONLY, or opt into live updates with ALWAYS_FETCH or CACHE_TTL.

• Keep prompts consistent and auditable across deployments by pairing fetch policies with the prompts lockfile and materialized directory.

---

   # Always fetch from API, use materialized as fallback
    prompt = langwatch.prompt.get_prompt(
        "customer-support-bot",
        fetch_policy=FetchPolicy.ALWAYS_FETCH
    )

    # Fetch every 5 minutes, use materialized between fetches
    prompt = langwatch.prompt.get_prompt(
        "customer-support-bot",
        fetch_policy=FetchPolicy.CACHE_TTL
        cache_ttl_minutes=5
    )

    # Never fetch, use materialized only (air-gapped mode)
    prompt = langwatch.prompt.get_prompt(
        "customer-support-bot",
        fetch_policy=FetchPolicy.MATERIALIZED_ONLY
    )

    # Default behavior: use materialized if available, otherwise fetch
    prompt = langwatch.prompt.get_prompt(
        "customer-support-bot",
        fetch_policy=FetchPolicy.MATERIALIZED_FIRST  # default
    )

Rollout

• Start: 2025-12-16

• Completion: 2025-12-16

• Availability: Python SDK, Typescript SDK

Notes

• https://docs.langwatch.ai/prompt-management/features/advanced/guaranteed-availability

---

Author & Review

• Author: @Andrew Joia

• Reviewed by: [@Reviewer]

• Created: 2025-12-16

---