baksohyeon
UserRun one AI agent as the orchestrator of your whole workspace, inside the Orca ADE, on the CLI subscription you already pay for. No API key, no model endpoint. Orca hosts and places every session; the harness dispatches workers under contract, verifies their output before accepting it, and hands the role to a fresh session when context runs out.
Categories
Indexed Skills (5)
delivery-gate
Stop hook that blocks Claude from finishing until quality checks pass. Detects rationalization patterns (surface text heuristics), stale learning logs (filesystem mtime), and low disk space. Complements self-audit by mechanically enforcing learning capture habits.
blame-agent
Structured incident observation tool for agent mistakes. Recovers what was observed, what was not observed, and where observation was promoted to cause too early, with [관측]/[강제]/[형성]/[판단] split. Use when: /blame-agent, "시말서", "roast your mistake", "왜 이거 망함", agent blame.
daily-triage
Read-only L1 daily triage loop. Sweeps default-branch commits, open PRs/MRs, code-review-graph signals, and stale memory/wiki, then writes a structured one-line-item report to the loop state file. Use when running the daily-triage loop or when the user asks for a repo triage or health sweep.
postmortem
Write a junior-friendly incident or postmortem note with evidence, hypotheses, and decision rationale. Use for postmortems, incident write-ups, RCA docs, or debugging retrospectives.
scope
Judge work size and recommend an operating unit (direct/quick/thread/phase/ADR). Use when the user asks to scope work, or when a natural-language request needs a scale/location recommendation before execution.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.