muratcankoylan

Ambiguous tools cause routing errors, malformed calls, and failed recovery. This skill provides practical patterns for clear schemas, descriptions, responses, and tool catalogs.

LLM projects often fail because task fit, architecture, costs, and output contracts remain unclear. This skill provides a staged method and reusable pipeline template.

Agent memory designs often add complexity without improving retrieval quality. This skill guides architecture choices and provides practical vector, graph, and temporal prototypes.

Multi-agent designs often add cost and coordination failures without improving outcomes. This skill helps select topologies, handoffs, consensus methods, and recovery controls.

Agent quality is difficult to measure because outputs vary and may have several valid forms. This skill builds repeatable rubrics, tests, gates, and monitoring.

Long AI sessions waste tokens and lose important context. This skill provides measured strategies for budgeting, masking, compaction, caching, retrieval, and partitioning.

Long contexts can hide critical instructions, preserve bad claims, and mix conflicting tasks. This skill diagnoses the failure pattern and recommends placement, filtering, compression, isolation, or recovery strategies.

Large context windows can still produce weak results when important information competes for attention. This skill explains how to curate focused context using practical mental models.

Long agent sessions can lose decisions, file history, and next steps during compression. This skill provides structured methods and probes that preserve operational context.

Agent decisions become difficult to explain when beliefs, goals, and commitments lack formal structure. This skill models BDI chains with RDF, SPARQL, provenance, and temporal validity.

LLM quality reviews often drift because judges use vague criteria and exhibit systematic bias. This skill provides calibrated rubrics, comparison methods, and validation metrics.