GUIDA alle applicazioni

Mercati di prompt e raccolte di prompt pubbliche

Public prompt hubs can help teams discover and adapt prompt examples, but SDK workflows may load more than visible instruction text.

Ascolta questa guidaNarrato dall'IA · ~3 min

In questa pagina3 minuti di lettura
  1. Panoramica
  2. Immersione profonda
  3. Implementazione nel mondo reale
  4. The Future of Prompt Marketplaces and Public Prompt Collections
  5. Metterlo in pratica
  6. Continua a esplorare
  7. Domande frequenti

Panoramica

LangSmith public prompt pulls can deserialize manifests containing model or prompt configuration; a 2026 security advisory documents risks including request redirection and exposure of prompt contents or credentials. Treat public manifests as untrusted executable configuration until reviewed, pinned, and tested.

Immersione profonda

Public prompt collections are useful for finding task structures and examples, but a prompt pulled through an SDK may include a serialized manifest with model configuration or LangChain objects. LangSmith’s documentation describes its public hub as community-created and warns that entries are user-generated, unverified, and not reviewed or endorsed. That statement applies to this hub; it should not be generalized to every prompt marketplace. More importantly, copying visible text and pulling a public manifest into an application are different trust decisions. The LangSmith SDK maintainers published GitHub advisory GHSA-3644-q5cj-c5c7 for public prompt pulls that deserialize untrusted manifests without an explicit trust-boundary warning. The advisory explains that a manifest may configure model endpoints, headers, or other constructor arguments, and may deserialize prompt or runnable objects. In affected workflows, a malicious public manifest could redirect requests to an attacker-controlled base URL or proxy and expose prompt contents, retrieved context, credentials, or other request data. The vulnerability applies when an application pulls a public owner/name prompt, the source is untrusted or compromised, and the application uses the manifest without independent review. It is not a claim that simply viewing a prompt in the web hub causes this impact. The advisory lists patched versions: Python langsmith 0.8.0 and later, JavaScript/TypeScript langsmith 0.6.0 and later, langchain 0.3.30 and later, and langchain-classic 1.0.7 and later. Patched SDKs block public owner/name pulls by default; callers must explicitly opt in with dangerously_pull_public_prompt or dangerouslyPullPublicPrompt. Do not treat that opt-in as a trust check. Review the manifest and source, pin an approved commit rather than relying on a moving latest reference, avoid include_model and secrets_from_env for untrusted sources, and keep credentials scoped. Apply code review, version control, testing, and audit practices as you would for executable configuration. Same-organization prompts also need access controls and review if credentials or accounts could be compromised.

Implementazione nel mondo reale

A developer pulls a public LangSmith prompt by owner/name and reviews its manifest, model settings, and pinned commit before considering it for an application.

A security reviewer rejects an untrusted manifest that configures an attacker-controlled model base URL or proxy that could redirect LLM traffic.

A team checks SDK versions against the LangSmith advisory and upgrades Python langsmith to 0.8.0 or later or JS/TS langsmith to 0.6.0 or later.

A prompt library process records the source commit, review evidence, and approved changes, then uses controlled credentials and regression tests before deployment.

The Future of Prompt Marketplaces and Public Prompt Collections

Prompt libraries may become more tightly integrated with application code and model configuration, which makes source provenance and change review increasingly important. SDK defaults and advisory guidance can change, so teams should track current maintained documentation and patch notices. A pinned commit, reviewed manifest, controlled credentials, and regression checks provide clearer control than a popularity score or a mutable “latest” reference. Public prompts should remain untrusted until a reviewer approves the exact version and use. Internal prompt libraries also need access control and change history to reduce risks from compromised accounts.

Metterlo in pratica

Perché è importante

  • Scelte di build. La progettazione a livello di applicazione determina se l’intelligenza artificiale migliora i risultati reali.
  • Team e flusso di lavoro. Una buona integrazione del flusso di lavoro crea guadagni di produttività di cui gli utenti possono fidarsi.
  • Rischio e sicurezza. I casi d'uso ben definiti riducono l'affaticamento dovuto al cambiamento e il rischio di implementazione.

Attenzione

  • Automatizzare un processo interrotto può amplificare i problemi esistenti.
  • I team potrebbero automatizzare eccessivamente e rimuovere il necessario giudizio umano.
  • La qualità può variare se i risultati non vengono valutati continuamente.

Prossimi passi

  1. Mappa il flusso di lavoro corrente e identifica la fase di maggiore attrito.
  2. Definisci checkpoint umani prima dell'automazione completa.
  3. Formare gli utenti su prompt, percorsi di escalation e standard di qualità.
  4. Tieni traccia dei risultati a livello di attività per confermare il valore duraturo.

Continua a esplorare

Newsletter gratuita

Ricevi il briefing quotidiano sull'IA

Tre storie verificate sull'IA ogni mattina nei giorni feriali, scritte in un inglese semplice. Gratuito per sempre, senza pubblicità.

Un'email ogni giorno feriale. Disiscriviti con un clic. Non vendiamo né condividiamo mai il tuo indirizzo.

Mettiti alla prova

Take the Prompt Marketplaces and Public Prompt Collections quiz

Feedback immediato su ogni risposta e un certificato condivisibile con un ID verificabile una volta superato un corso.

Inizia il quiz

Sostieni l'istruzione gratuita sull'IA. AI Understanding è un'organizzazione non profit 501(c)(3) — nessun annuncio, nessun paywall, mai. Fai una donazione

Domande frequenti

What is Prompt Marketplaces and Public Prompt Collections?

Public prompt hubs can help teams discover and adapt prompt examples, but SDK workflows may load more than visible instruction text. LangSmith public prompt pulls can deserialize manifests containing model or prompt configuration; a 2026 security advisory documents risks including request redirection and exposure of prompt contents or credentials. Treat public manifests as untrusted executable configuration until reviewed, pinned, and tested.

What does LangChain say about prompts in its public LangSmith hub?

The current docs give this warning specifically for the public hub.

Why can an SDK pull involve more risk than copying visible prompt text?

The advisory says public manifests are deserialized and may configure runtime objects.

Which impact does GHSA-3644-q5cj-c5c7 describe?

The advisory describes request redirection and possible disclosure when vulnerable applications pull untrusted manifests.

Which Python langsmith version is listed as the patched floor in the advisory?

The advisory lists langsmith Python versions before 0.8.0 as affected and 0.8.0 as patched.

Which JavaScript/TypeScript langsmith version is listed as patched?

The advisory lists langsmith npm versions before 0.6.0 as affected and 0.6.0 as patched.