Ramsbaby/openclaw-self-healing
Skill Claude CodeCodex
Five-level autonomous self-healing system for OpenClaw Gateway with preflight config validation, persistent learning, reasoning logs, and multi-channel alerts. Features Claude Code as the Level 3 emergency doctor for AI-powered diagnosis and repair.
Plugin Claude Code
Workflow skills for AI-pair-programming and autonomous agent loops, with a focus on bias removal, irreversibility guards, and convergent requirement gating. English public bodies live in plugins/; the maintainer's Korean-persona overlays stay in private/ (gitignored).
Skill Claude CodeCodex
An interactive setup wizard for the Jarvis Discord bot, a program that runs inside Discord. It collects required settings, configures updates, and checks optional local search support.
Skill Claude CodeCodex
A skill for saving confirmed facts, decisions, preferences, and important project context in the Jarvis wiki. The same stored information can be shared across its different interfaces.
MCP server Claude CodeCodexCursor +2
MCP server "serena" as configured in Ramsbaby/jarvis. Runs locally from the git+https://github.com/oraios/serena Python package. Needs 1 environment variable to run.
MCP server Claude CodeCodexCursor +2
MCP server "serena-board" as configured in Ramsbaby/jarvis. Runs locally from the git+https://github.com/oraios/serena Python package. Needs 1 environment variable to run.
Instructions file
Instructions for Ramsbaby/jarvis, covering ai navigation (start here), development rules, golden path — 작업 착수 전 고정 순서 (2026-04-21), surface memory boundary — 표면 통합 메모리 경계 and 사용 규칙.
Command
A command set for switching between saved company and personal Claude accounts and checking their status. It uses a local script and can synchronize login data between the system keychain and a file.
Command
A command for turning a complex request into a step-by-step work plan before carrying it out. It is intended for multi-step tasks, refactoring, and large migrations.
Command
A command for comparing models, prompts, or approaches by running measurements on sample tasks. It produces sample-level results and a recommendation based on the comparison.
Command
A standard procedure for coding work that covers exploring the code, planning, editing, checking the result, and reporting what changed. It is intended for new features, refactoring, and bug fixes.
Command
A command for managing a coding session when its conversation history becomes large. It explains when to compact the context, start a new session, split work, or use another agent.
Command
A crisis-response procedure for software incidents such as an outage, a stopped bot, a critical bug, or a changed external dependency. It is written in Korean.
Command
A read-only security-audit command based on OWASP and STRIDE, which are common methods for finding web and software security risks. It traces realistic attack paths through areas such as authentication, payments, data, infrastructure, and code.
Command
A deployment command for updating the Jarvis system and checking that its services still run. Deployment means applying code or configuration changes to a running system.
Command
A visual review command for finished HTML, CSS, SVG, images, documents, and interface screenshots. It checks for design patterns that can make an output look generic or obviously machine-made.
Command
A command for tracing the underlying cause of failures and reproducible bugs using a structured chain of questions. It is designed for incidents such as failed scheduled jobs, stopped bots, and unresponsive services.
Command
A command that saves reusable discoveries and lessons from a coding session as one-line JSON records. JSONL means one JSON object per line, making the file easy to append to and process.
Command
A command for switching between predefined MCP tool profiles for OpenClaw infrastructure, a Flutter/Dart app, or a Gatsby/React blog. MCP is a way of connecting an agent to external tools and services.
Command
A decision-making conversation that lays out at least three options, their trade-offs, and one recommendation. It also notes whether each choice can be undone, while leaving the final decision to you.
Command
A command for immediately adding a documented mistake to a shared Markdown file. It records the recurring pattern, what happened, verifiable evidence, and a prevention step.
Command
A command that splits a complex task into three parallel roles: research, coding, and validation. Each role handles a separate part of the work before the results are combined.
Command
A command for reviewing a feature plan, architecture document, or technical proposal through an 11-section checklist. Architecture describes how a system is organized and operates.
Command
A retrospective command that summarizes what went well, what went wrong, and what should change after a work session. It can record recurring mistakes in an error notebook using the fields pattern, actual case, evidence, and response.