Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By fyrsmithlabs
Cross-session memory and learning for Claude Code. Provides semantic search, memory recording with category classification, session-level granularity, checkpoints, error remediation, conversation indexing, self-reflection, and multi-agent orchestration via the contextd MCP server (requires contextd v0.4.0+).
npx claudepluginhub fyrsmithlabs/marketplace --plugin contextdSave a checkpoint of the current session state including accomplishments, in-progress work, and next steps. Use before clearing context or ending a session.
Diagnose an error message using AI analysis and past fixes. Searches remediations and provides root cause analysis with recommended next steps.
List all available skills and commands for contextd Claude Code plugin.
Execute multi-task orchestration from GitHub issues or epics with parallel agents and consensus reviews
Analyze memories and remediations for behavior patterns, evaluate policy compliance, and optionally remediate with pressure-tested doc updates. Use for self-improvement and quality assurance.
Multi-agent workflow orchestrator using context-folding for parallel sub-agent execution, ReasoningBank for learning capture, and checkpoints for progress tracking. Use for complex tasks requiring multiple specialized agents.
Unified agent for debugging, refactoring, architecture analysis, and general contextd-enforced work. Automatically detects task mode based on keywords and applies the appropriate workflow.
Use when executing multi-task work from GitHub issues or epics with parallel agents, dependency management, and consensus reviews. Triggers on "orchestrate issues", "execute epic", "run the orchestration", or complex multi-feature development phases. REQUIRES contextd MCP server.
Use when reviewing agent behavior patterns, improving CLAUDE.md based on past failures, or checking ReasoningBank health. REQUIRES contextd MCP server - this skill is inoperable without it.
Use when onboarding to a project or creating/updating CLAUDE.md - covers codebase analysis, documentation generation, and policy management
Use when starting any session with contextd - introduces core tools for cross-session memory, semantic code search, and error remediation. REQUIRES contextd MCP server.
Use for session lifecycle management - covers session start/end protocols, checkpoint workflow, and error remediation flow
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Context management with save and restore capabilities
Agent enforcement framework — context injection, planning gates, session learning
Persistent context for AI coding assistants
(Alpha) Persistent memory, architectural decisions, and safety guardrails for Claude Code. Your agent starts every session with full project context — stack, decisions, patterns, safety rules, and a handoff from the previous session.
Persistent memory system for Claude Code with enforced context
Memory compression system for Claude Code - persist context across sessions
Claude Code plugins for fyrsmithlabs project standards and workflows. Includes YAGNI/KISS enforcement, git-repo-standards, git-workflows with multi-agent consensus review, and GitHub planning integration.
Terminal Elegance design system compliance checking for FyrsmithLabs products. Audits CSS, templates, and documentation for hardcoded values, accessibility issues, and brand consistency.
A Claude Code plugin marketplace providing skills, commands, and agents for fyrsmithlabs project standards and workflows.
Before installing, ensure you have:
gh) 2.0+ - Required for planning features (install)Add the marketplace to your Claude Code settings:
# Install via Claude Code plugin marketplace
claude plugin install fyrsmithlabs/marketplace
# Initialize a project with standards
/fs-dev:init
# Run a brainstorm session for a new feature
/fs-dev:brainstorm "Add user authentication"
# Check your daily standup
/fs-dev:standup
# Run multi-agent code review
/fs-dev:consensus-review
This marketplace contains three plugins that can be installed independently:
| Plugin | Version | Components | Purpose |
|---|---|---|---|
| fs-dev | 1.9.1 | 13 skills, 15 agents, 12 commands | Development standards, workflows, planning |
| contextd | 1.2.0 | 5 skills, 2 agents, 8 commands | Cross-session memory via MCP |
| fs-design | 1.0.0 | 1 skill, 2 agents, 1 command | Design system compliance |
Set up a project to follow fyrsmithlabs standards with interactive wizard, auto-detection, and compliance checking.
# Full interactive wizard
/fs-dev:init
# Audit compliance without making changes
/fs-dev:init --check
# Quick setup (skip wizard)
/fs-dev:init --quick
# Check for drift from standards
/fs-dev:init --validate
Interactive design interview with complexity assessment, requirements gathering, and GitHub planning.
# Start a brainstorm session
/fs-dev:brainstorm "Add OAuth support"
# Brainstorm creates GitHub Issues automatically in Phase 6
Run parallel code review with 6 specialized agents. All agents have veto power by default. Features adaptive token budgets, context-folding isolation for large scopes, and progressive summarization.
# Review staged changes
/fs-dev:consensus-review
# Review specific path
/fs-dev:consensus-review src/auth/
# Ignore veto power (advisory only)
/fs-dev:consensus-review --ignore-vetos
# Run subset of agents
/fs-dev:consensus-review --agents security,code-quality,go
Save your session state for later resumption. Essential before clearing context or ending a session.
# Save current session state
/contextd:checkpoint
# Checkpoints capture:
# - Accomplishments
# - In-progress work
# - Next steps
Audit CSS, templates, and documentation for Terminal Elegance design system compliance.
# Scan default locations
/fs-design:check
# Check specific file
/fs-design:check static/css/main.css
# Check directory
/fs-design:check internal/templates/
# 1. Design the feature
/fs-dev:brainstorm "Add user authentication"
# 2. Create GitHub issues (automatic from brainstorm Phase 6)
# or manually:
/fs-dev:plan "Add user authentication"
# 3. Code implementation...
# 4. Multi-agent code review
/fs-dev:consensus-review
# 5. Commit with standards
git commit -m "feat: add user authentication"
# 1. Check current state
/contextd:status
# 2. Search for relevant past learnings
/contextd:search authentication patterns
# 3. Work on tasks...
# 4. Save progress before clearing context
/contextd:checkpoint
# 5. Record what you learned
/contextd:remember
# Full codebase analysis
/fs-dev:discover
# Security-focused analysis
/fs-dev:discover --lens security
# Performance-focused analysis
/fs-dev:discover --lens performance
# Today's status with GitHub PR/Issue synthesis
/fs-dev:standup
# Weekly summary
/fs-dev:standup --week