Conversation
| @@ -121,9 +121,6 @@ A Helm chart for kagent, built with Google ADK | |||
| | database.postgres.url | string | `"postgres://postgres:kagent@pgsql-postgresql.kagent.svc.cluster.local:5432/postgres"` | | | |||
There was a problem hiding this comment.
There are additional fields as part of the work to remove sqlite. See the values file in the PR.
| - The `database.type` configuration option has been removed. | ||
| - SQLite-related Helm values (`database.sqlite.*`) have been removed. | ||
| - A PostgreSQL instance with pgvector is now bundled in the Helm chart by default. | ||
| - The bundled PostgreSQL is deployed automatically when both `database.postgres.url` and `database.postgres.urlFile` are empty. |
There was a problem hiding this comment.
Probably worth mentioning the additional settable fields for configuring the bundled postgres image, resources, username/password, etc. See the values file in the PR.
…ntation Add 17 new MDX pages covering: - Core Concepts: tools ecosystem, human-in-the-loop, agent memory, prompt templates, multi-runtime (Go/Python ADK), git-based skills, context management - Resources: full Agent CRD reference (Agent, ModelConfig, RemoteMCPServer, MCPServer, ModelProviderConfig) - Solo Enterprise: overview, installation, security & access control, multicluster federation, observability (OTel/ClickHouse), enterprise UI, BYO agent frameworks (ADK/CrewAI/LangGraph), model providers, tools & MCP Update navigation.json with entries for all new pages. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> Signed-off-by: Art Berger <art.berger@solo.io>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> Signed-off-by: Art Berger <art.berger@solo.io>
1. All tool names wrong — kubectl_* prefix should be k8s_*, many tools fabricated or misnamed 2. Memory config YAML wrong — uses nonexistent enabled/embeddingProvider fields; should be modelConfig/ttlDays 3. Context management YAML wrong — eventsCompaction.enabled should be compaction with compactionInterval/etc. 4. Prompt template dataSources wrong — configMapRef.name should be flat kind/name/alias fields 5. BYO agent YAML nesting wrong — image/replicas should be under byo.deployment, not at byo level 6. Memory API endpoints wrong — uses path params but actual API uses query params, plus missing endpoints 7. OCI skills refs format wrong — refs is a string array, not objects with image key 8. Read-only mode (kagentTools.readOnly) — no evidence this exists 9. navigation.json indentation bug — line 292 lost leading whitespace Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Signed-off-by: Art Berger <art.berger@solo.io>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Signed-off-by: Art Berger <art.berger@solo.io>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Signed-off-by: Art Berger <art.berger@solo.io>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Signed-off-by: Art Berger <art.berger@solo.io>
… items, namespace in YAML Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Signed-off-by: Art Berger <art.berger@solo.io>
…section Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Signed-off-by: Art Berger <art.berger@solo.io>
the rendering of the codeblock did not work Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art Berger <art.berger@solo.io>
- Concepts: add cards for Tools Ecosystem, Human-in-the-Loop, Agent Memory, Prompt Templates, Multi-Runtime, Git-Based Skills, Context Management - Getting Started: add Local Development card - Examples: add Agents via MCP, BYO CrewAI, BYO LangGraph cards - Resources: fix Helm chart href (/helm/helm -> /helm) Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art Berger <art.berger@solo.io>
Related to kagent-dev/kagent#1502 Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Yaniv Marom Nachumi <yanivmar@amdocs.com> Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Yaniv Marom Nachumi <yanivmar@amdocs.com> Signed-off-by: Art Berger <art.berger@solo.io>
683226e to
0922925
Compare
Signed-off-by: Art Berger <art.berger@solo.io>
kristin-kronstain-brown
left a comment
There was a problem hiding this comment.
I had lots of questions but I didn't see anything glaringly wrong.
| - **Searchable.** Agents retrieve relevant memories via cosine similarity. | ||
| - **Automatic.** Agents extract and save user intent, key learnings, and preferences without explicit user action. | ||
| - **Time-bounded.** Memories expire after a configurable TTL (default 15 days). | ||
| - **Shared storage.** Memory uses the same kagent database (PostgreSQL), not a separate database. |
There was a problem hiding this comment.
| - **Shared storage.** Memory uses the same kagent database (PostgreSQL), not a separate database. | |
| - **Shared storage.** Memory uses the kagent database (PostgreSQL), not a separate database. |
I don't know why, but with same, I want it to say something like "Memory uses the same database as kagent..." Like with "same" I want an "as". Maybe a preference rather than a requirement.
|
|
||
| To enable memory, set the `memory` field on the declarative agent spec. The `modelConfig` field references a `ModelConfig` object whose embedding provider generates memory vectors. | ||
|
|
||
| You can also configure memory in the kagent UI when you create or edit an agent by selecting an embedding model and setting the memory TTL. |
There was a problem hiding this comment.
Should these be steps? It's not an overly complicated sentence but there is an "or" and an "and" on something that could potentially be steps.
There was a problem hiding this comment.
Or are you saying that by running this YAML, then you can change these settings in the UI?
| |------|-------------| | ||
| | `save_memory` | Explicitly save a piece of information. | | ||
| | `load_memory` | Search for relevant memories by query. | | ||
| | `prefetch_memory` | Automatically run before response generation. | |
There was a problem hiding this comment.
Run what exactly? The agent?
|
|
||
| ### Viewing Memories | ||
|
|
||
| In the kagent UI, you can view the memories that an agent has saved. This lets you inspect what the agent has learned and retained from past interactions. |
There was a problem hiding this comment.
| In the kagent UI, you can view the memories that an agent has saved. This lets you inspect what the agent has learned and retained from past interactions. | |
| In the kagent UI, you can view the memories that an agent has saved. With saved memories, you can inspect what the agent has learned and retained from past interactions. |
|
|
||
| ## Limitations | ||
|
|
||
| - **No per-memory deletion.** You can clear all memories for an agent, but you cannot delete individual memory entries. |
There was a problem hiding this comment.
Does clear mean the same thing as delete?
|
|
||
| The `kagent-builtin-prompts` ConfigMap ships with five reusable templates: `skills-usage`, `tool-usage-best-practices`, `safety-guardrails`, `kubernetes-context`, and `a2a-communication`. | ||
|
|
||
| For more information, see the [Agents](/docs/kagent/concepts/agents#prompt-templates) concept guide. |
There was a problem hiding this comment.
| For more information, see the [Agents](/docs/kagent/concepts/agents#prompt-templates) concept guide. | |
| For more information, see [Agents](/docs/kagent/concepts/agents#prompt-templates). |
|
|
||
| Long conversations can now be automatically compacted to stay within LLM context windows. You can configure the `context.compaction` field to enable periodic summarization of older events while preserving key information. | ||
|
|
||
| For more information, see the [Agents](/docs/kagent/concepts/agents#context-management) concept guide. |
There was a problem hiding this comment.
| For more information, see the [Agents](/docs/kagent/concepts/agents#context-management) concept guide. | |
| For more information, see [Agents](/docs/kagent/concepts/agents#context-management). |
Does the same link need be included in every paragraph?
|
|
||
| **Migration:** | ||
|
|
||
| If you were using the default SQLite backend, no migration is needed. The bundled PostgreSQL will be deployed automatically. You can optionally customize the bundled instance via `database.postgres.bundled.*` (storage size, image, credentials) as needed. See the [Database configuration guide](/operations/operational-considerations/#database-configuration) for details. |
There was a problem hiding this comment.
| If you were using the default SQLite backend, no migration is needed. The bundled PostgreSQL will be deployed automatically. You can optionally customize the bundled instance via `database.postgres.bundled.*` (storage size, image, credentials) as needed. See the [Database configuration guide](/operations/operational-considerations/#database-configuration) for details. | |
| If you were using the default SQLite backend, no migration is needed. The bundled PostgreSQL is deployed automatically. You can optionally customize the bundled instance via `database.postgres.bundled.*` (storage size, image, credentials) as needed. See the [Database configuration guide](/operations/operational-considerations/#database-configuration) for details. |
|
|
||
| If you were using the default SQLite backend, no migration is needed. The bundled PostgreSQL will be deployed automatically. You can optionally customize the bundled instance via `database.postgres.bundled.*` (storage size, image, credentials) as needed. See the [Database configuration guide](/operations/operational-considerations/#database-configuration) for details. | ||
|
|
||
| Note that for production deployments, it is recommended to use your own external PostgreSQL instance. If you already are, you can keep your `database.postgres.url` or `database.postgres.urlFile` settings as before the migration. |
There was a problem hiding this comment.
| Note that for production deployments, it is recommended to use your own external PostgreSQL instance. If you already are, you can keep your `database.postgres.url` or `database.postgres.urlFile` settings as before the migration. | |
| Note that for production deployments, use your own external PostgreSQL instance. If you already are, you can keep your `database.postgres.url` or `database.postgres.urlFile` settings as before the migration. |
|
|
||
| | Tool | Description | | ||
| |------|-------------| | ||
| | `k8s_get_resources` | Query resources (pods, deployments, services, etc.). | |
There was a problem hiding this comment.
| | `k8s_get_resources` | Query resources (pods, deployments, services, etc.). | | |
| | `k8s_get_resources` | Query resources, such as pods, deployments, and services. | |
Signed-off-by: Art Berger <art.berger@solo.io>
https: //github.com/kagent-dev/kagent/pull/1527 Signed-off-by: Art Berger <art.berger@solo.io>
Signed-off-by: Art <artberger@users.noreply.github.com>
Follow up to #321, #332
Adds concepts, guides, and release notes for new 0.8 features, including: