Hermes Agent Skills & Knowledge Base

Skills, worldview, and knowledge base for the Hermes Agent autonomous orchestration engine.

Structure

• worldview.md — Philosophy of good agency (distilled from 2600 lessons)
• skills/ — Individual skill files, one per skill

Skills

Skill	Description
core-principles	Distilled behavioral principles for AI assistant operation
foundational-mindset	2600 core principles organized by SOUL.md layers
worldview	Philosophy of good agency — not rules, but a worldview to reason from
browser-art-deployment	Workflow for deploying HTML/p5.js creative projects
config-change-verify	Verify model/server configuration changes with a live cursor
creative-visualizer-debug	Troubleshooting and deploying creative visualizers
dogfood	Systematic exploratory QA testing of web applications
github-monitor-pattern	GitHub monitoring pattern for issue/PR tracking
gpu-task-isolation	Protocol for running heavy GPU workloads on specific GPUs
llama-cpp-memory-workaround	Workaround for llama.cpp memory save HTTP 500 errors
muscle-run-wrapper-protocol	Protocol for executing high-compute media-heavy tasks
plan	Plan mode for Hermes — inspect context, write markdown plan
rsi-logging-protocol	RSI logging protocol
safe-file-editing	Protocol for safely modifying existing files
subagent-driven-development	Using subagents for independent implementation
systematic-debugging	Systematic debugging methodology
test-driven-development	TDD workflow for features and bugfixes
test-skill	Test skill for verifying RSI loop detection
whisper-lyric-gen	OpenAI Whisper word-level timestamped JS generation
whisper-vocal-isolation-debug	Troubleshooting Whisper missing early/quiet vocals