Shared agent scaffolding with skills, knowledge bundles, and cross-agent context for AI agents (Goose, Hermes, Claude Code, etc.)
Skill Claude CodeCodex
Fix the warning Failed to update project tracker with instruction: Failed to parse projects.json file that can occur after updating Goose CLI, caused by malformed JSON in the project tracker file.
Skill Claude CodeCodex
Fix the Goose Desktop shell tool so that shell commands run with the full user environment — including devbox/nix-installed tools, crc/oc, cargo, sdkman, and custom exports.
Skill Claude CodeCodex
Create and configure a Goose custom provider that points to the local Headroom context-optimization proxy. This skill covers only the Goose provider configuration — for Headroom installation and systemd setup, see the headroom-litellm-proxy skill.
Skill Claude Code
Set up Goose (CLI and Desktop) to use a local LiteLLM proxy as a custom provider. This covers the RedHat provider pattern — a local LiteLLM proxy backed by Vertex AI Claude models (no auth, http://localhost:4000).
Skill Claude CodeCodex
Set up Goose (CLI and Desktop) to use a remote MaaS endpoint — a shared LiteLLM instance that requires API key authentication (e.g., https://maas-rhdp.apps.maas.redhatworkshops.io).
Skill Claude Code
Manage the next-ai-draw-io MCP server as a Goose custom extension. This skill installs the @next-ai-drawio/mcp-server npm package locally (avoiding repeated npx downloads), registers it as a stdio extension in Goose's config.yaml, and provides teardown and upgrade operations. Use when the user wants AI-powered draw.io…
Skill Claude CodeCodex
Identify and remove Goose session records from the sessions database. Covers all four session categories with per-category opt-in and optional date filtering.
Skill Claude CodeCodex
Configure Goose with persistent instructions that apply across all projects and sessions.
Skill Claude Code
Configure Goose (CLI and Desktop) to use a remote GPU-hosted LLM model as a custom provider named Skupper. The model is served through a Skupper V2 Virtual Application Network managed by skupper-model-provider, which exposes remote GPU hosts (rhel-ai, rhtevan-work) as local endpoints on localhost:9000 and…
Skill Claude CodeCodex
Install the Headroom context-optimization proxy and run it as a systemd user-scope service, chained to a local LiteLLM proxy for upstream LLM access.
Skill Claude CodeCodex
Check the health, configuration, and runtime status of the local Headroom context-optimization proxy.
Skill Claude CodeCodex
Configure Hermes Agent so it discovers AgentFS skills, making the same skill library available to both Goose and Hermes without duplication.
Skill Claude CodeCodex
Symptom: bash: warning: setlocale: LCCTYPE: cannot change locale (UTF-8): No such file or directory — appears 6× in the Desktop in-app terminal on every shell start. Root cause: Electron inherits macOS-style LCCTYPE=UTF-8 (bare, no language prefix like enUS.). Bash calls setlocale() during startup before any…
Skill Claude Code
Configure Hermes Agent to use the local Headroom context-optimization proxy as its custom LLM provider, with model discovery and verification.
Skill Claude Code
Set up Hermes Agent (CLI and Desktop) to use a locally hosted LiteLLM proxy as its custom LLM provider. The proxy is backed by Vertex AI Claude models (no auth, http://127.0.0.1:4000).
Skill Claude Code
Deploy, manage, and test self-hosted LLM model containers on NVIDIA GPUs via Podman. All deployments use profiles — curated, VRAM-validated configurations. No individual model aliases.
Skill Claude CodeCodex
Check the health, configuration, and runtime status of the local LiteLLM proxy.
Skill Claude Code
Set up a local LiteLLM proxy that fronts GCP Vertex AI as an OpenAI-compatible endpoint, managed by systemd, with proper service account credentials.
Skill Claude CodeCodex
Extract knowledge from the current chat session and persist it as an OKF v0.1 knowledge bundle under /.agents/knowledge/.
Skill Claude Code
Harvest generalizable concepts from project-scoped MEMORY.md files and distill them into OKF-compliant knowledge bundles under /.agents/knowledge/.
Skill Claude CodeCodex
Audit and repair index.md files throughout an Open Knowledge Format (OKF) v0.1 knowledge bundle tree.
Skill Claude CodeCodex needs its repo
Create, organize, or update a directory as an Open Knowledge Format (OKF) v0.1 knowledge bundle.
Skill Claude Code needs its repo
Create new skills, modify and improve existing ones with built-in AgentFS conventions. Operates in three modes: simple (quick scaffold), advanced (full eval/iterate/optimize loop using upstream Anthropic skill-creator), and skill check (audit existing skills against five quality principles). Default mode is simple…
Skill Claude Code
Harvest procedural knowledge from project-scoped MEMORY.md files and distill them into reusable skills under /.agents/skills/.
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: