Agent Skills by Floris Vossebeld
Showing 28 vetted skills indexed across 2 GitHub repositories.
FVossebeldcontext-engineering-collection
Bundle 24 context-engineering skills with Azure-native bindings for agent systems.
FVossebeldskill-template
Create new Skill units with standardized SKILL.md frontmatter and sections.
FVossebeldfoundry-tool-governance
Design and govern Microsoft Foundry tool surfaces with curated, versioned, policy-controlled toolboxes.
FVossebeldagent-publishing
Publish Microsoft-native agents as versioned, governance-bound production endpoints.
FVossebeldfabric-data-agent
Route analytics queries to Fabric semantic models with governance rules.
FVossebeldazure-identity-for-agents
Bind agent access to Microsoft Entra identities for credential-free SDK use and RBAC boundaries.
FVossebeldfoundry-iq-knowledge-layer
Centralize enterprise knowledge with permission-aware grounding across SharePoint, data stores, and web content.
FVossebeldazure-agentic-retrieval
Enable permission-aware retrieval and grounding across Azure AI Search, File Search, and SharePoint.
FVossebeldazure-memory-state
Select Azure-native storage backends for durable agent memory management.
FVossebeldresponsible-ai-guardrails
Enforce layered guardrails for prompts, tool usage, and outputs in Microsoft-native agent contexts.
FVossebeldreasoning-trace-optimizer
Analyze reasoning traces to detect failure patterns and optimize prompts.
FVossebelddigital-brain
Organize personal knowledge and content workflows with modular, memory-backed agent frameworks.
FVossebeldbook-sft-pipeline
Convert books into SFT datasets and orchestrate LoRA training pipelines.
FVossebeldcomprehensive-research-agent
Coordinate multi-source research with structured thinking and source tracking.
FVossebeldmulti-agent-patterns
Coordinate multi-agent systems with supervisor routing, handoffs, and failure handling.
FVossebeldcontext-degradation
Diagnose context degradation and generate structured recovery reports for agent sessions.
FVossebeldcontext-compression
Summarize long conversations into anchored sections to reduce token usage.
FVossebeldmemory-systems
Store and retrieve structured memories with vector and graph search.
FVossebeldadvanced-evaluation
Automate LLM output evaluation with scoring, pairwise comparisons, and bias mitigation.
FVossebeldharness-engineering
Design autonomous agent harnesses with surface classifications, durable logs, and governance gates.
FVossebeldcontext-fundamentals
Explain context engineering fundamentals including attention mechanics and context window anatomy.
FVossebeldproject-development
Codify project-level decision-making for LLM-powered batch pipelines and workflows.
FVossebeldevaluation
Evaluate agent outputs with weighted rubric criteria and per-dimension scores.
FVossebeldcontext-optimization
Reduce token usage by masking, compacting, caching, and partitioning context.