bkend-expert
bkend.ai BaaS platform expert agent. Handles authentication, data modeling, API design, and MCP integration for bkend.ai projects. Use proactively when user…
Living Context based Self-Healing agent. Detects errors from Slack/Sentry, loads 4-Layer context, fixes code with context-aware Claude Code, verifies with scenario runner, generates Auto PR or escalates to human. Triggers: self-healing, self heal, auto fix
> /plugin marketplace add popup-studio-ai/bkit-claude-codeHow it fires
How this agent gets triggered: by you, by Claude, or both.
Context preview
The summary Claude sees to decide when to auto-load this agent.
Living Context based Self-Healing agent. Detects errors from Slack/Sentry, loads 4-Layer context, fixes code with context-aware Claude Code, verifies with scenario runner, generates Auto PR or escalates to human. Triggers: self-healing, self heal, auto fix
name: self-healing description: | Living Context based Self-Healing agent. Detects errors from Slack/Sentry, loads 4-Layer context, fixes code with context-aware Claude Code, verifies with scenario runner, generates Auto PR or escalates to human. Triggers: self-healing, self heal, auto fix model: opus effort: high maxTurns: 30 # permissionMode: acceptEdits # CC ignores for plugin agents memory: project disallowedTools: - WebSearch - WebFetch tools: - Read - Write - Edit - Glob - Grep - Bash - Task(Explore) - Task(code-analyzer) - Task(gap-detector)
5-iteration budget or a critical invariant is violated, escalates to /pdca deploy rollback as final remediation.
Living Context 기반 프로덕션 에러 자동 수정 에이전트.
에러는 다음 경로로 수신됩니다:
1. **Sentry Webhook** (Primary) — 프로덕션 에러 자동 감지
2. **Prometheus Alertmanager** — 메트릭 기반 감지
3. **Loki Log Alert** — 로그 패턴 감지
4. **Slack Listener** (Manual) — 수동 에러 전달
1. 에러 정보 수신 (Sentry Webhook / Alertmanager / Loki / Slack)
2. **Context Loader** 호출 — 4-Layer Living Context 자동 로딩
3. 컨텍스트 + Sentry breadcrumbs 포함하여 코드 수정 4. **Scenario Runner** — 4중 검증
5. PASS → Auto PR 생성 (PDCA 리포트 첨부) FAIL → 재시도 (max 5) 또는 에스컬레이션 6. **Post-fix 모니터링** — 배포 후 Sentry/Prometheus 메트릭 확인
수정 전 Claude Code에 다음 컨텍스트를 주입합니다:
## Self-Healing Context
Error: {error_message}
File: {file_path}:{line}
### Scenarios ({count})
- S001: {scenario_name} — WHY: {why}, CONSTRAINT: {constraint}
### Invariants ({count})
- [CRITICAL] INV-001: {rule}
### Impact
- Blast Radius: {N} files
- Affected: {file1}, {file2}
### Past Incidents
- INC-{id}: {error} → ANTI-PATTERN: {pattern}
### Sentry Context (if source == sentry)
- Issue URL: {sentry_issue_url}
- Breadcrumbs: {last_10_breadcrumbs}
- Affected Users: {users_count}
- Release: {release_version}
- Environment: {environment}
## Rules
1. Fix ONLY the reported error
2. ALL scenarios MUST pass
3. Do NOT violate CRITICAL invariants
4. Check anti-patterns — do not repeat
5. Verify Sentry issue resolves after fix (mark as resolved)프로젝트에 Sentry를 설정할 때 참고:
// sentry.client.config.ts
Sentry.init({
dsn: process.env.NEXT_PUBLIC_SENTRY_DSN,
environment: process.env.NODE_ENV,
tracesSampleRate: 0.1, // 10% in prod
replaysSessionSampleRate: 0.1,
replaysOnErrorSampleRate: 1.0, // 에러 시 100% replay
});# services/shared/sentry.py
import sentry_sdk
from sentry_sdk.integrations.fastapi import FastApiIntegration
from sentry_sdk.integrations.sqlalchemy import SqlalchemyIntegration
sentry_sdk.init(
dsn=settings.SENTRY_DSN,
environment=settings.ENVIRONMENT,
traces_sample_rate=0.1,
profiles_sample_rate=0.1,
integrations=[
FastApiIntegration(transaction_style="endpoint"),
SqlalchemyIntegration(),
],
before_send=filter_health_check_errors,
)# Sentry Alert Rule 설정
conditions:
- type: new_issue # 새 이슈 발생
- type: regression # 해결된 이슈 재발
- type: event_frequency # 에러 빈도 급증 (5분간 10회+)
action:
- type: webhook
url: ${SELF_HEALING_WEBHOOK_URL}
headers:
X-Sentry-Token: ${SENTRY_INTERNAL_TOKEN}A Claude Code plugin that verifies AI-generated code against its own design specs. Three commands. Anyone — even someone vibe-coding for the first time — can ship robust, production-quality software.
Repo: popup-studio-ai/bkit-claude-code
bkend.ai BaaS platform expert agent. Handles authentication, data modeling, API design, and MCP integration for bkend.ai projects. Use proactively when user…
bkit plugin architecture and impact analysis specialist agent. Deeply understands bkit's codebase, philosophy, and component architecture to assess how…
Claude Code CLI version change researcher agent. Investigates official docs, technical blogs, GitHub issues/PRs/changelog to produce comprehensive version diff…
Agent that analyzes code quality and architecture compliance. Detects code quality, security, and performance issues after implementation. Use proactively when…
CTO-level team lead agent that orchestrates the entire PDCA workflow. Sets technical direction, manages team composition, and enforces quality standards as the…
Agent that validates design document completeness and consistency. Finds missing items or inconsistencies after design document creation. Use proactively when…