davsclaus opened a new pull request, #1834:
URL: https://github.com/apache/camel-website/pull/1834
## What this changes
Four commits, each can be reviewed on its own.
**1. Outdated facts on the home page and in llms.txt**
- Home page: 280+ releases (282 release tags; the old 300+ counted RC and
milestone tags), export to Maven only (Gradle export was removed in 4.19), and
Camel Kafka Connector runs Kamelets as Kafka Connect connectors (same text on
/projects/ and /docs/).
- Footer: add the Camel CLI, write "Camel K", and link the /latest/ docs
like the rest of the site.
- llms.txt:
- The Camel TUI is a developer tool, as its manual says, not for
monitoring integrations in production.
- Two broken links are fixed (the EIP index and OpenTelemetry, now
opentelemetry2).
- The llm: component works with any OpenAI-compatible API.
- Small facts are corrected: median bug fix time of 1 day, the benchmark
machine is a 64 GB Mac mini, most examples (not all) have a Citrus test, the
core team share is 78-96% since 2009, Claus Ibsen is a committer since 2008,
and only content pages have a Markdown version.
**2. Installer manifests for 4.23.0, 4.22.2, 4.18.5, 4.18.4 and 4.14.9**
The install command on the home page still installed 4.22.1.
`latest.properties` now points to 4.23.0. The manifests were generated with
`update-camel-cli-manifest.sh`, but from the `camel-4.23.0` tag instead of
main, because main's `install.sh` now says "Java 21+ required" (for 4.24). At
the tag, `install.sh`, `install.ps1` and the generator are byte-identical to
the files already on the site, so only the manifests change.
**3. llms.txt: AI features from Camel 4.22 and 4.23**
- A new section on semantic decisions with System One models:
- the Semantic language, with Jev from TypeSafe AI and Wolf Defender as
experts
- audit
- version notes for 4.23 and 4.24
- a YAML example that validates with both versions
- AI tools: the MCP Server component is Stable since 4.22, tools can return
structured output, the AI Resource component exists, and
camel-mcp-server-starter works on Spring Boot.
- AI components:
- the LangChain4j Agent component, and models configured with properties
only
- what the llm: component covers
- the Spring AI 2.0 components
- Also: `camel run --observe`, the documentation examples validated since
4.23, and the AGENTS.md file that `camel export` and the archetypes generate.
- One entry for the four local model benchmark posts.
**4. The Semantic language example in the 4.23 release post**
It had no expert and put the instructions directly on the evaluation. It
fails validation on main and would fail at startup on 4.23. It now follows the
documented syntax.
## How it was checked
- `node --test test/*-test.js`: 23 tests pass. The changed data files parse
as YAML.
- Every new or changed link returns 200 on camel.apache.org.
- The semantic YAML examples pass `camel validate yaml` on main and with
`--camel-version=4.23.0`.
- The facts were checked against the sources:
- the Camel catalog and CLI source
- the 4.19 upgrade guide
- the TUI manual
- the GitHub API (282 release tags)
- Maven Central (camel-launcher 4.22.2 and 4.23.0 are published)
Not changed: the Camel Kit links. The move to apache/camel-kit is in
progress.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
https://claude.ai/code/session_01STT6whBgK1AqsSsUKrnE8m
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
