melikhanmutlu/agent-manager-skill
skills-extra/agent-manager-skill/SKILL.md
Manage multiple local CLI agents via tmux sessions (start/stop/monitor/assign) with cron-friendly scheduling. Includes agent performance optimization workflows.
tools
Updated Jun 7, 2026
$ install --global
skillsauthnpx skillsauth add melikhanmutlu/web_ar agent-manager-skillInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Jun 7, 2026, 6:04 AM49.9s1 file scanned
SKILL.md
- name:
- agent-manager-skill
- description:
- Manage multiple local CLI agents via tmux sessions (start/stop/monitor/assign) with cron-friendly scheduling. Includes agent performance optimization workflows.
- risk:
- unknown
- source:
- community
Agent Manager Skill
When to use
Use this skill when you need to:
- run multiple local CLI agents in parallel (separate tmux sessions)
- start/stop agents and tail their logs
- assign tasks to agents and monitor output
- schedule recurring agent work (cron)
Prerequisites
Install agent-manager-skill in your workspace:
git clone https://github.com/fractalmind-ai/agent-manager-skill.git
Common commands
python3 agent-manager/scripts/main.py doctor
python3 agent-manager/scripts/main.py list
python3 agent-manager/scripts/main.py start EMP_0001
python3 agent-manager/scripts/main.py monitor EMP_0001 --follow
python3 agent-manager/scripts/main.py assign EMP_0002 <<'EOF'
Follow teams/fractalmind-ai-maintenance.md Workflow
EOF
Notes
- Requires
tmuxandpython3. - Agents are configured under an
agents/directory (see the repo for examples).
Agent Performance Optimization Workflow
Systematic improvement of existing agents through performance analysis, prompt engineering, and continuous iteration.
Phase 1: Performance Analysis and Baseline
Gather Performance Data
Collect metrics including:
- Task completion rate (successful vs failed tasks)
- Response accuracy and factual correctness
- Tool usage efficiency (correct tools, call frequency)
- Average response time and token consumption
- User satisfaction indicators (corrections, retries)
- Hallucination incidents and error patterns
User Feedback Pattern Analysis
Identify recurring patterns:
- Correction patterns: Where users consistently modify outputs
- Clarification requests: Common areas of ambiguity
- Task abandonment: Points where users give up
- Follow-up questions: Indicators of incomplete responses
- Positive feedback: Successful patterns to preserve
Failure Mode Classification
Categorize failures by root cause:
- Instruction misunderstanding, output format errors
- Context loss in long conversations
- Tool misuse or inefficient tool selection
- Constraint violations, edge case handling
Baseline Performance Report
- Task Success Rate: [X%]
- Average Corrections per Task: [Y]
- Tool Call Efficiency: [Z%]
- User Satisfaction Score: [1-10]
- Average Response Latency: [Xms]
- Token Efficiency Ratio: [X:Y]
Phase 2: Prompt Engineering Improvements
- Chain-of-Thought Enhancement: Add explicit reasoning steps and self-verification checkpoints
- Few-Shot Example Optimization: Curate diverse examples covering common use cases and edge cases
- Role Definition Refinement: Clear mission, expertise domains, behavioral traits, constraints
- Constitutional AI Integration: Self-correction with critique-and-revise loops
- Output Format Tuning: Structured templates, dynamic formatting, progressive disclosure
Phase 3: Testing and Validation
A/B Testing Framework
- Minimum sample size: 100 tasks per variant
- Confidence level: 95% (p < 0.05)
- Compare: success rate, speed, token usage
- Use blind human review + automated scoring
Evaluation Metrics
- Task-Level: Completion rate, correctness, efficiency, tool appropriateness
- Quality: Hallucination rate, consistency, format compliance, safety
- Performance: Latency, token consumption, cost per task
Phase 4: Version Control and Deployment
Version Management
Format: agent-name-v[MAJOR].[MINOR].[PATCH]
MAJOR: Significant capability changes
MINOR: Prompt improvements, new examples
PATCH: Bug fixes, minor adjustments
Staged Rollout
- Alpha (5% traffic) -> Beta (20%) -> Canary (20-50-100%) -> Full deployment
- 7-day monitoring period after full deployment
Rollback Triggers
- Success rate drops >10% from baseline
- Critical errors increase >5%
- Cost per task increases >20%
- Safety violations detected
Success Criteria
- Task success rate improves by >=15%
- User corrections decrease by >=25%
- No increase in safety violations
- Response time remains within 10% of baseline
Continuous Improvement Cycle
- Weekly: Monitor metrics and collect feedback
- Monthly: Analyze patterns and plan improvements
- Quarterly: Major version updates with new capabilities
Related Skills
melikhanmutlu/skills/market
tools
VerifiedTrustedCommunity
# AI Marketing Suite — Main Orchestrator You are a comprehensive AI marketing analysis and content generation system for Claude Code. You help entrepreneurs, agency builders, and solopreneurs analyze websites, generate marketing content, audit funnels, create client proposals, and build marketing strategies — all from the command line. ## Command Reference | Command | Description | Output | |---------|-------------|--------| | `/market audit <url>` | Full marketing audit (parallel subagents)
SKILL.mdUpdated Jun 30, 2026
melikhanmutlu/skills/market-social
testing
VerifiedTrustedCommunity
# Social Media Content Calendar & Generation You are the social media engine for `/market social <topic/url>`. You generate a complete 30-day content calendar with platform-specific posts, hooks, hashtags, and a content repurposing strategy. Every post is ready to publish or hand to a social media manager. ## When This Skill Is Invoked The user runs `/market social <topic/url>`. If a URL is provided, fetch the site to understand the brand, audience, and content themes. If a topic is provided,
SKILL.mdUpdated Jun 30, 2026
melikhanmutlu/skills/market-seo
development
VerifiedTrustedCommunity
# SEO Content Audit ## Skill Purpose Perform a comprehensive SEO audit of a webpage or website, covering on-page SEO, content quality (E-E-A-T), keyword analysis, technical SEO, and content strategy. This skill combines automated analysis via `scripts/analyze_page.py` with expert-level manual review to produce an actionable SEO audit document. ## When to Use - User provides a URL and asks for SEO analysis, audit, or recommendations - User wants to improve organic search rankings and traffic -
SKILL.mdUpdated Jun 30, 2026
melikhanmutlu/skills/market-report
tools
VerifiedTrustedCommunity
# Marketing Report Generator (Markdown Format) ## Skill Purpose Generate a comprehensive, professionally formatted marketing report in Markdown. This skill compiles data from all previous audit and analysis results into a single, client-ready document with scores, findings, recommendations, and a prioritized action plan with revenue impact estimates. ## When to Use - User wants a full marketing report for a client or their own business - User has completed one or more audit skills and wants a
SKILL.mdUpdated Jun 30, 2026