A Claude Code plugin marketplace. Context-continuity and workflow-survival plugins.
Add the SurvivalDev marketplace:
claude plugin marketplace add B-A-M-N/SurvivalDevPluginsInstall any plugin:
claude plugin install charmed@survivaldev-plugins
claude plugin install handshake@survivaldev-plugins
claude plugin install vibeflow@survivaldev-plugins
claude plugin install survivaldev-infradoctor@survivaldev-plugins
claude plugin install converge@survivaldev-plugins
claude plugin install prforge@survivaldev-pluginsCheck installed plugins:
claude plugin list| Plugin | Source | Description |
|---|---|---|
| charmed | B-A-M-N/charmed |
Charm ecosystem TUI intelligence — architectural reasoning, anti-pattern detection, intent-aware scaffolding, and UX critique for Bubble Tea / Lip Gloss / Bubbles projects |
| handshake | B-A-M-N/handshake |
Context-continuity — checkpoints workflow state before compaction/clear and rehydrates after |
| survivaldev-infradoctor | B-A-M-N/survivaldev-infradoctor |
Infrastructure doctor for self-hosted Firecrawl and SearXNG |
| vibeflow | B-A-M-N/vibeflow |
Mistral Vibe workflow design, implementation, and validation lifecycle |
| converge | B-A-M-N/converge |
Universal recurring task engine — cron, loops, background tasks |
| prforge | B-A-M-N/prforge |
Professional PR contribution harness with guarded release and mesh mode |
Charm ecosystem TUI intelligence — architectural reasoning, anti-pattern detection, intent-aware scaffolding, and UX critique for Bubble Tea / Lip Gloss / Bubbles projects.
- 6 skills: tui-audit, tui-scaffold, tui-refactor, tui-explain, tui-instrument, tui-ux-review
- MCP server (
charmed-mcp): project scanning, knowledge base access, DeepWiki references - ATree integration (Unity-Lab-AI/ATree): code graph enrichment for model detection, message flow, and impact analysis
- Knowledge base: embedded archetypes, constraints, timing models, and component ontology — no external API calls
- Triggers from natural language: "audit my TUI", "scaffold a terminal app", "what is viewport.Model?"
Prevents agent memory loss during compaction, /clear, or long-running sessions. Automatically checkpoints workflow state before context reset and rehydrates after.
- Hook-driven:
PreCompact,SessionStart,UserPromptSubmit - Skill-driven toggle:
/handshake on|off|status - Checkpoint format: structured JSON (machine) + Markdown (human)
Hook-driven diagnostics and repair for self-hosted Firecrawl and SearXNG instances. Detects failures, diagnoses root causes, applies safe repairs, and validates recovery.
- Autonomous diagnosis → propose fix → safe repair → validate → report loop
- Dispatched automatically on infrastructure incidents
Single-model Claude Code plugin for conceptualizing, designing, planning, applying, and validating advanced Mistral Vibe workflows. Uses signed intent artifacts, progressive feasibility references, Mermaid/JSON runtime diagrams, and deterministic validation gates.
- Lifecycle:
init → design → plan → apply → validate - Init emits signed
VISION.md,PLAN.md, andWORKFLOW_CONTRACT.json - Design maps intent to feasible runtime surfaces instead of assuming impossible middleware/tool/session behavior
- Feasibility substrate covers tiers A-E, key contracts, workflow patterns, events, config keys, tier composition, and observability
Automates recurring tasks, scheduled workflows, and loop-based execution for AI agents running in CLI environments.
- Cron-based scheduling with natural language support
- Loop execution with self-paced cadence
- Background task management and wake-up calls
Delegated PR execution with guarded release, coding discipline enforcement, and distributed mesh mode. Turns the agent into a disciplined upstream contributor.
- Model: investigate → plan → patch → validate → self-review → package → you approve
- Enforces repo intelligence, scope contracts, validation honesty, and git safety
- Distributed mesh mode for multi-repo contribution workflows
Browse all plugins by topic: claude-code plugin survivaldev