Mb Reliability — Development skill for Claude Code
Show which ModelBound skills work in real use and which need attention.
How to install Mb Reliability
Installs to ~/.claude/skills/modelbound-modelbound-claude-code-plugin-mb-reliability/SKILL.md
mkdir -p ~/.claude/skills/modelbound-modelbound-claude-code-plugin-mb-reliability && curl -fsSL https://raw.githubusercontent.com/ModelBound/modelbound-claude-code-plugin/HEAD/commands/mb-reliability.md -o ~/.claude/skills/modelbound-modelbound-claude-code-plugin-mb-reliability/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Mb Reliability does
description: Show which ModelBound skills work in real use and which need attention argument-hint: [days]
mb reliability --days "${1:-30}"
Shows per-skill reliability (worked ÷ reported) over the window, with failure and partial counts. Skills with fewer than three reports are shown as "too few to score" rather than a misleading percentage.
Use it before picking a skill for an important task, and after applying a suggested fix to confirm the fix held.
Alternatives in Development
- Resume Work From A Handoff Document — You are tasked with resuming work from a handoff document through an interactive process 10k ★
- Lab Skills — Skills distilled from the Lab's real work and collaboration practices 309 ★
- Innovate — Draft N novel sparse-attention submissions in one shot — algorithm-innovation focus, all must compile, no benc 68 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Agent Session Status
Live OpenCode, Claude Code, and Codex session state for Ironbar and Waybar. Provider events are reduced to a s
Agentchirp
A macOS menu bar app that tells you when your Claude Code agents need attention — without you having to go loo
Pilovo
Chrome extension that shows whether your ChatGPT and Claude tabs are working, have a response ready or need yo
Dash Status
Show submodule, registry, drift, and attention status for the dash
Evidence And Tabs
How a real pentester uses Burp's surfaces — Proxy history search, Repeater tabs, Intruder, Organizer, annotati
Load Handoff
If the user invokes as 'load-handoff latest' then just load the most recent HANDOFF file in the project's hand
Related Agents
Heavy Executor
Execute DEMANDING tasks that need real reasoning or design judgment (intelligence/taste need 7-8) - real root-
Cob
Full-capability Sonnet programmer. Use when you need to delegate implementation work without sacrificing quali
Engineering Kael
Use this agent when you need technical leadership for feature development, architecture decisions, implementat