Muratcan Koylan
Community@muratcankoylan · Toronto
I just enjoy building stuff with AI, for AI.
Agent Skills by Muratcan Koylan
Showing 25 vetted skills indexed across 2 GitHub repositories.
muratcankoylancontext-engineering-collection
Curates context engineering skills for building and evaluating production AI agent systems.
muratcankoylanharness-engineering
Design control surfaces, feedback loops, and governance boundaries for autonomous agent systems.
muratcankoylanself-improvement-loops
Designs self-modifying agent loops that mine failures and evolve harnesses under acceptance gates.
muratcankoylanlatent-briefing
Compact orchestrator trajectories into worker KV caches using Attention Matching for multi-agent memory sharing.
muratcankoylanlong-horizon-prompting
Write pseudo-formal task briefs for long-running autonomous agents and parallel multi-agent orchestrations.
muratcankoylanBACKPRESSURE
Create deterministic self-verification checks for AI coding agents.
muratcankoylancontext-engineering-collection
Centralizes best practices for AI agent context engineering.
muratcankoylanskill-template
Create new Skill units with a standardized SKILL.md template and YAML frontmatter.
muratcankoylandigital-brain
Manage a modular personal knowledge system with append-only JSONL, YAML, and MD data.
muratcankoylanreasoning-trace-optimizer
Analyze interleaved thinking traces to detect failure patterns and optimize prompts.
muratcankoylanbook-sft-pipeline
Convert ePub books into JSONL SFT datasets and train LoRA style models.
muratcankoylancomprehensive-research-agent
Organizes source-driven research into a synthesized, citable report.
muratcankoylanmulti-agent-patterns
Coordinate multi-agent tasks with explicit handoffs and context isolation.
muratcankoylanbdi-mental-states
Translate agent cognition prompts into RDF-based BDI mental states linked to world states.
muratcankoylancontext-degradation
Diagnose context degradation patterns and provide mitigation strategies for agent systems.
muratcankoylancontext-compression
Summarize long-running agent sessions while tracking modified files and decisions.
muratcankoylanmemory-systems
Persist agent memory across sessions using multi-layer architecture and temporal knowledge graphs.
muratcankoylanadvanced-evaluation
Automate LLM output evaluation with direct scoring and pairwise comparisons.
muratcankoylancontext-fundamentals
Explain context components and progressive disclosure for AI agent systems.
muratcankoylanproject-development
Guide end-to-end LLM projects through a five-stage pipeline with structured outputs.
muratcankoylanevaluation
Evaluate agent performance across configurations and sessions with multi-dimensional rubrics.
muratcankoylancontext-optimization
Compacts, masks, caches, and partitions context to reduce token usage under budget management.
muratcankoylanfilesystem-context
Offload oversized context data to filesystem files with grep/read_file retrieval.
muratcankoylantool-design
Design tool interfaces with schemas, error handling, and naming conventions.