メニュー
Create a linked JIRA ticket for documentation work. No agent dispatch. Uses direct JIRA REST API calls via a shell script. Checks for existing links, handles public/private project visibility, converts markdown to JIRA wiki markup.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | docs-workflow-create-jira |
| description | Create a linked JIRA ticket for documentation work. No agent dispatch. Uses direct JIRA REST API calls via a shell script. Checks for existing links, handles public/private project visibility, converts markdown to JIRA wiki markup. |
| model | claude-haiku-4-5@20251001 |
| argument-hint | <ticket> --base-path <path> --project <PROJECT> |
| allowed-tools | Read, Write, Bash |
Step skill for the docs-orchestrator pipeline. Follows the step skill contract: parse args → do work → write output.
Unlike other step skills, this skill does not dispatch an agent. It runs scripts/create-jira-ticket.sh directly.
Output: null (produces a JIRA URL, not a file)
$1 — Parent JIRA ticket ID (required)--base-path <path> — Base output path (e.g., .agent_workspace/proj-123)--project <PROJECT> — Target JIRA project key for the new ticket (required)<base-path>/planning/plan.md
Requires JIRA_API_TOKEN (or the backward-compatible alias JIRA_AUTH_TOKEN) and JIRA_EMAIL in the environment. create-jira-ticket.sh sources ~/.env then <project-root>/.env, where the project root is resolved from the PLAN_FILE location.
Run the create-jira-ticket script:
bash ${CLAUDE_SKILL_DIR}/scripts/create-jira-ticket.sh "$TICKET" "$PROJECT" "${BASE_PATH}/planning/plan.md"
The script handles all steps:
scripts/md2wiki.py for markdown → wiki conversion[ccs] Docs - prefixThe script prints the JIRA URL on success (e.g., https://redhat.atlassian.net/browse/DOCS-456), or a skip message if a linked ticket already exists.
After the script completes, write the sidecar to ${BASE_PATH}/create-jira/step-result.json:
mkdir -p "${BASE_PATH}/create-jira"
If the script created a new ticket (output contains a JIRA URL like https://...atlassian.net/browse/DOCS-456):
{
"schema_version": 1,
"step": "create-jira",
"ticket": "<TICKET>",
"completed_at": "<current ISO 8601 timestamp>",
"jira_url": "https://redhat.atlassian.net/browse/DOCS-456",
"jira_key": "DOCS-456",
"action": "created",
"skipped": false,
"skip_reason": null
}
If the script found an existing linked ticket (output contains "already exists"):
{
"schema_version": 1,
"step": "create-jira",
"ticket": "<TICKET>",
"completed_at": "<current ISO 8601 timestamp>",
"jira_url": "https://redhat.atlassian.net/browse/<linked-key from output>",
"jira_key": "<linked-key from output>",
"action": "found_existing",
"skipped": false,
"skip_reason": null
}
Extract jira_key and jira_url from the script's stdout. If the script fails (non-zero exit), write the sidecar with action: "skipped", skipped: true, and skip_reason: "create_failed".
Documentation workflow orchestrator. Reads the step list from .agent_workspace/docs-workflow.yaml (or the plugin default). Runs steps sequentially, manages progress state, handles iteration and confirmation gates. Claude is the orchestrator — the YAML is a step list, not a workflow engine.
Write documentation from a documentation plan. Dispatches the docs-writer agent. Supports AsciiDoc (default) and MkDocs formats. Default placement is UPDATE-IN-PLACE; use --draft for staging area. Also supports fix mode for applying technical review corrections.
Score documentation quality and intent alignment using LLM judge agents (Opus). Dispatches two judge agents in parallel (doc_quality, intent_alignment), extracts specific gaps from AC coverage analysis, and cross-references against scope-req-audit evidence. Produces pass/fail gate with actionable gap list. Iteration logic is owned by the orchestrator, not this skill.
Run code-learner analysis on a source repository for the docs orchestrator workflow. Dispatches a subagent to run learn-code, keeping the heavy orchestration out of the main context. Produces ONBOARDING.md, module registry, per-module summaries, and cross-module relationship data.
Classify JIRA requirements by code evidence status before planning. Uses learn-code analysis data and source code inspection to determine if each requirement is grounded, partial, or absent. Fans out one subagent per requirement for isolated classification. Prevents hallucinated documentation for unimplemented features and surfaces gaps for implemented ones. Conditional on has_source_repo.
Technical accuracy review with optional code-learner claim validation. Iteration logic owned by the orchestrator.