Production-grade skills for Hermes Agent - self-evolving, persona-aware, slash-command native
Skill Claude CodeCodex
Use when the user needs DevOps or SRE help including CI/CD pipelines, Docker optimization, Kubernetes manifests, infrastructure safety, or production log analysis.
Skill Claude CodeCodex
Use when the user needs expert Python engineering help including code review, performance profiling, pytest generation, package scaffolding, or static type guidance.
Skill Claude CodeCodex
Use when the user presents a vague, ambiguous, or underspecified task. Conducts structured multi-turn dialogue to extract true requirements with 95% clarity before any code is written.
Skill Claude CodeCodex
Use when starting a new feature, project, or significant refactor. Writes a comprehensive PRD/spec document before any implementation code, following the "spec-first, code-second" principle.
Skill Claude CodeCodex
Use before starting any code generation task. Reads Hermes Agent's SOUL.md to dynamically adapt code style, comment tone, documentation voice, and architectural decisions to match the agent's defined persona.
Skill Claude CodeCodex
Use after completing a complex task, fixing a tricky error, or discovering a reusable workflow. Analyzes the execution record, proposes skill improvements, and feeds into Hermes Agent's self-learning loop — the meta-skill that makes all other skills better.
Skill Claude CodeCodex
Use when setting up or optimizing CI/CD pipelines. Generates GitHub Actions workflow configurations following best practices — matrix builds, caching, artifact management, and deployment gates.
Skill Claude CodeCodex
Use before merging or committing code. Performs six-axis code quality gating — security, complexity, style, test coverage, documentation, and dependency health. Blocks low-quality code from entering the main branch.
Skill Claude CodeCodex
Use when debugging complex issues involving multiple modalities (frontend + backend, API + UI, terminal + browser). Coordinates Hermes Agent's built-in debugger, browser, terminal, and vision tools for systematic root-cause analysis.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: