bootstrap
Meta-skill chaining /specify → /constitute → /plan → /tasks → spec-validation → dispatch-ready. Greenfield project from idea to…
Escalate a YouTrack issue to Claude Code (Tier 2) for CODE IMPLEMENTATION ONLY. Use ONLY when someone explicitly asks to implement, fix, refactor, build, or write code. Do NOT use for questions, lookups, or status checks — use youtrack-lookup for those instead.
$ npx -y skills add papadopouloskyriakos/agentic-chatops --skill escalate-to-claude --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/escalate-to-claudeContext preview
The summary Claude sees to decide when to auto-load this skill.
Escalate a YouTrack issue to Claude Code (Tier 2) for CODE IMPLEMENTATION ONLY. Use ONLY when someone explicitly asks to implement, fix, refactor, build, or write code. Do NOT use for questions, lookups, or status checks — use youtrack-lookup for those instead.
name: escalate-to-claude
description: Escalate a YouTrack issue to Claude Code (Tier 2) for CODE IMPLEMENTATION ONLY. Use ONLY when someone explicitly asks to implement, fix, refactor, build, or write code. Do NOT use for questions, lookups, or status checks — use youtrack-lookup for those instead.
allowed-tools: Bash
user-invocable: true
metadata:
openclaw:
always: trueYou are Tier 1. When a task requires code implementation, architecture decisions, or multi-file changes, you MUST escalate to Claude Code (Tier 2) using the escalation script.
Run this command:
./skills/escalate-to-claude.sh <ISSUE-ID>
The script: 1. Validates the issue ID format (PROJECT-NUMBER) 2. Fetches the summary from YouTrack if not provided 3. Moves the issue to "In Progress" in YouTrack 4. Fires the n8n webhook to start a Claude Code session
./skills/escalate-to-claude.sh CUBEOS-31
./skills/escalate-to-claude.sh CUBEOS-31 "SMAZ2 compression implementation"
1. **ALWAYS run the script.** Do NOT just say "I've escalated" — you MUST execute `./skills/escalate-to-claude.sh`. The script output will confirm "OK: Escalated" or show an error. 2. **Do NOT provide implementation details.** If someone asks to implement something, do NOT give code, analysis, pros/cons, or partial answers. Just escalate. 3. **Extract the issue ID** from the conversation. If no issue ID is mentioned, ask the user which issue, or create one first with `./skills/yt-create-issue.sh CUBEOS "summary" "description"` then escalate the new ID. 4. **Confirm with the script output.** After running, tell the user: "Escalated ISSUE-ID to Claude Code. Session starting." and include the HTTP status from the script output. 5. **One escalation per issue.** If already escalated in this conversation, say so instead of re-running.
AI agents that triage infrastructure alerts, investigate root causes, and propose fixes — while a solo operator sleeps. For the complete technical reference, see README.extensive.md.
Meta-skill chaining /specify → /constitute → /plan → /tasks → spec-validation → dispatch-ready. Greenfield project from idea to…
Phase C of greenfield project bootstrap. Produce constitution.md (5-15 non-negotiable articles) + steering/*.md (cross-cutting…
Phase D of greenfield project bootstrap. Produce design.md + data-model.md + contracts/{openapi,asyncapi,schemas/*} +…
Phase B of greenfield project bootstrap. Interview the operator to produce PROJECT.json + PROJECT.md + spec/NNN-*/requirements.md…
Phase E of greenfield project bootstrap. Atomize design.md + contracts into tasks.json with files_owned + dependencies +…
Query code graph database for function callers, callees, dependencies, and dead code in CubeOS and MeshSat repos.