| name | project-red-hat-doc-skill-authoring |
| metadata | {"author":"rhoai3-demo","version":"1.0.0","platform-family":"red-hat","platform-baseline":"repo","ocp-baseline":"repo","skill-group":"Project Structure"} |
| description | Create or update a compliant product skill from official Red Hat documentation for this demo's platform stack. Use when the user provides Red Hat OpenShift AI, OpenShift Container Platform, or OpenShift Data Foundation documentation sources and asks to create rhoai-*, ocp-*, or odf-* skills; when defining a new Red Hat product skill family; when capturing source evidence, extraction rules, validation criteria, examples, or a component roadmap; or when deciding whether a source belongs in RHOAI, OCP, ODF, project, or environment guidance. Do NOT use for live cluster deployment or troubleshooting; use env-* skills. Do NOT use as product authority by itself; generated skills must cite official Red Hat docs and verification commands.
|
Red Hat Doc Skill Authoring
Use this meta-skill to convert official Red Hat product documentation into
durable repo skills. The output is usually a new or updated component skill
under .agents/skills/ with one of these prefixes:
rhoai-* for Red Hat OpenShift AI component knowledge
ocp-* for Red Hat OpenShift Container Platform component knowledge
odf-* for Red Hat OpenShift Data Foundation component knowledge
Required References
Read only the files needed for the current task:
.agents/references/red-hat-doc-map.yaml for routing Red Hat product
documentation sources and repo routing categories to flat repo skills
references/platform-skill-taxonomy.md for product-family routing,
frontmatter metadata, and rule/roadmap expectations
references/source-capture.md for the source ledger template and baseline
gate
references/extraction-rules.md for what to extract from official docs
references/validation-checklist.md before finishing the generated skill
examples/component-skill-outline.md when creating a new component skill
Workflow
- Read
docs/PLATFORM_BASELINE.md.
- Read
.agents/references/red-hat-doc-map.yaml and identify whether the
official source already maps to an existing, planned, or blocked skill.
- Identify the product family and prefix:
- Red Hat OpenShift AI Self-Managed ->
rhoai-*
- Red Hat OpenShift Container Platform ->
ocp-*
- Red Hat OpenShift Data Foundation ->
odf-*
- Confirm the input source matches the active baseline for that product.
If the product version is not pinned in
docs/PLATFORM_BASELINE.md, stop
after drafting the source-capture note and ask to pin the baseline before
creating a reusable component skill.
- If the doc-map route is
active, update the mapped skill unless the source
clearly belongs to a different capability. If the route is planned, create
the planned flat skill. If the route is missing, choose a stable flat skill
name and add a route for it. If the route is blocked-baseline, do not
create the reusable skill until the baseline is pinned.
- Capture official sources before writing the skill:
- product and version
- chapter or page URL
- source type, such as versioned docs.redhat.com documentation, Customer
Portal article, external Red Hat product docs, or repo-specific guidance
- documentation category
- retrieved date
- sections used
- related Red Hat docs, articles, and rh-brain examples
- Extract only supported product behavior:
- purpose and supported posture
- prerequisites
- install/configure/use/upgrade/remove workflows
- operators, namespaces, subscriptions, CRDs, API groups, fields, labels,
annotations, images, and artifacts
- validation commands and failure signals
- boundaries between Red Hat product behavior and demo glue
- Design the target skill:
- folder name uses the selected prefix and a stable capability name
- frontmatter
name matches the folder name
- metadata uses
platform-baseline: "repo" and ocp-baseline: "repo"
metadata.platform-family is rhoai, ocp, or odf
metadata.skill-group matches the taxonomy
- description lists strong positive triggers and negative triggers
.agents/references/red-hat-doc-map.yaml
AGENTS.md
.agents/rules/<family>.md when that family has at least one real skill
.agents/skills/project-structure/SKILL.md
.agents/skills/project-agent-guidance/SKILL.md
- the relevant roadmap under
.agents/skills/project-structure/references/
Output Contract
A completed generated component skill should normally contain:
.agents/skills/<prefix>-<capability>/
SKILL.md
references/
source-capture.md
official-doc-extraction.md
validation-checklist.md
examples/
<minimal-example>.md or focused manifest/runbook snippets
Keep generated examples minimal. Prefer snippets that demonstrate exact field
placement, verification commands, or recommended workflow shape.
Hard Rules
- Do not invent CR fields, API versions, annotations, operator channels, image
names, storage classes, installation modes, support posture, or recommended
settings.
- Do not treat upstream community documentation as Red Hat product authority.
- Do not treat rh-brain as product configuration truth; use it for narrative,
Red Hat articles, blogs, and code examples after official docs are captured.
- Do not create nested skill folders to mirror Red Hat documentation
categories; represent product/category/book/chapter routing and source type
in
.agents/references/red-hat-doc-map.yaml.
- Do not generate new
odf-* skills unless the ODF product version and
official documentation source are pinned in docs/PLATFORM_BASELINE.md.
- If official docs are ambiguous, mark the item unresolved and include an
oc explain, CRD, CSV, catalog, registry, or image lookup command.
- If the official source cannot be accessed and no local export is provided,
create only a draft scaffold or stop and ask for source content.
References
references/platform-skill-taxonomy.md
references/source-capture.md
references/extraction-rules.md
references/validation-checklist.md
examples/component-skill-outline.md
.agents/references/red-hat-doc-map.yaml