Component
core/ai-tool/retrieval
A read-only tool that fetches reference material by similarity, so an agent can ground an answer in the documents its store holds rather than only in what the model already holds. It declares a required socket onto a datastore whose engine serves similarity search — vector, semantic-memory or search — so the store is resolved in the topology rather than configured in a prompt. The developer sets how many matches a lookup returns, defaulting to five. Its effect is pinned read-only in its own schema. Realized on Kubernetes as a Knative Service, which admits an inbound call while the tool is running no replicas, so the tool occupies no compute between the calls that invoke it, and on Docker Compose as one service that runs while the stack runs. It publishes a host, port, endpoint and protocol, and a consumer wires to those without naming how the tool runs. The catalog pins no image, so nothing renders until the Release sets `infra.image`. The store's address reaches the tool's code under default variable names; an `inject:` block on the connection delivers instead, as variables or files, the names that code reads. The component's own `inject:` block takes `env` and `file` entries too, and one reading a value of the tool alone replaces that value's default variable name.
Versione
Provenienza
- Digest del contenuto
- sha256:7ef1482e831edd8428703ffdd39a1bae5092b0c8912737f5df94e41daf851d5b
- Pubblicato
- 2026-11-15T08:08:58Z
- Identità di firma
- https://token.actions.githubusercontent.com/https://github.com/aggrega-ai/aggrega-hub/.github/workflows/catalog-publish-sign.yml@refs/heads/main
Firma
- Ufficialeapplication/vnd.dev.sigstore.bundle.v0.3+json
https://token.actions.githubusercontent.com/https://github.com/aggrega-ai/aggrega-hub/.github/workflows/catalog-publish-sign.yml@refs/heads/main
sha256:b7ee3c99e90b0cef9ef78aa4bea3b0ee531f25de58326e923fa55e69d71972e2
Riferimento a questa versione
oci://hub.aggrega.ai/component/core/ai-tool/retrieval:1.0.1Parametri dello sviluppatore
Valori che lo sviluppatore imposta nell’Aggregate.
Estende BaseComponents#/schemas/BaseToolParameter
| Nome | Tipo | Predefinito | Obbligatorio | Descrizione |
|---|---|---|---|---|
| top_k | integer | 5 | sì | — |
| side_effects | string (read_only) | read_only | sì | — |
Dimensionamento dell’operatore
Dimensionamento dell’infrastruttura che l’operatore regola per ogni ambiente.
| Nome | Tipo | Predefinito | Obbligatorio | Descrizione |
|---|---|---|---|---|
| image | string | — | no | Container image serving the lookup over HTTP. |
| port | integer (min 1024, max 65535) | 8080 | no | Port the tool's own container listens on. A consumer does not necessarily connect here: the realization publishes the address callers use. On Kubernetes the control plane fronting the tool routes from its own port to this one; on Docker Compose callers reach this port directly. |
| compute | object | — | no | — |
| min_replicas | integer (min 0) | 0 | no | Replica floor. On Kubernetes, zero is what lets the tool run no replicas between calls, which the realization reaches through its control plane. Docker Compose runs one container throughout and refuses a floor above one. |
| max_replicas | integer (min 1) | 5 | no | Replica ceiling the Kubernetes realization's autoscaler may reach. On Docker Compose the one container never passes it. |
| control_plane_present | boolean | false | no | Operator affirmation that Knative Serving, the control plane the Kubernetes realization needs, is installed in the target cluster. That realization halts the build unless this is set true, so a custom resource is never emitted as an orphan into a cluster that cannot reconcile it. Docker Compose involves no control plane and does not read it. |
| overrides | object (open) | — | no | Realization-specific overrides. Accepts any nested structure. Each Recipe documents which keys it reads. |
Contratto di output
Valori che questo componente pubblica verso i consumatori attraverso la topologia uses.
| Nome | Tipo | Predefinito | Obbligatorio | Descrizione |
|---|---|---|---|---|
| host | string | — | sì | Name an agent calls to fetch reference material by similarity: the tool's in-cluster DNS name on Kubernetes, its service name on the Compose network on Docker Compose. A consumer that composes its own address dials this name; one that does not reads `endpoint` instead. |
| port | integer | — | sì | TCP port the lookup is accepted on. On Kubernetes the realization fronts the tool on the control plane's own port rather than the tool container's, so a consumer reads this rather than the port an operator set; on Docker Compose it is infra.port. |
| endpoint | string | — | sì | Full base URL an agent posts a lookup to, with the host and the port already composed. It addresses the tool, not the datastore behind it, which the tool reaches over its own socket. |
| protocol | string | — | no | URL scheme to reach `host` and `port` over, for a consumer composing its own address rather than reading `endpoint`. |
Socket uses
I socket di dipendenza attraverso cui questo componente si connette. La destinazione di ogni connessione collegata attraverso un socket deve soddisfare il suo selettore di etichette, e ogni connessione può portare un aggancio negli slot che il socket dichiara. Uno slot obbligatorio deve essere riempito da ogni connessione che l’Aggregate o la Release collega attraverso il socket.
target_datastore
- Selettore
- datastore.aggrega.ai/engine in (vector, semantic-memory, search)
- Iniezione
- env, file
Questo socket non dichiara slot di aggancio, quindi una connessione attraverso di esso non porta alcun aggancio.
Slot di aggancio
Punti di associazione con nome in cui una Feature si aggancia a questo componente. Il selettore di uno slot limita quale Feature può riempirlo e la sua lista inject indica come può essere consegnato il materiale della Feature. Uno slot obbligatorio deve essere associato su ogni istanza, da un aggancio sull’istanza o a un ambito superiore, e uno slot sensibile è dichiarato per gestire materiale segreto. Una Feature può anche agganciarsi con un nome che nessuno slot dichiara.
Nessuno slot di aggancio dichiarato.
Elenco consentito di auto-iniezione
env, file
Relazioni
Archi tipizzati da e verso altre entità pubblicate. Un riferimento senza una destinazione pubblicata viene mostrato come arco interrotto anziché come link.
Usa
- usausa il socket «target_datastore»
datastore.aggrega.ai/engine in (vector, semantic-memory, search)
Non dichiara slot di aggancio.
- $refestende lo schema (values)
BaseComponents#/schemas/BaseToolParameter
Usato da
si applica a
implementa
implementa