A curated collection of reusable agent skills for AI assistants, designed for easy redistribution and integration.
Agent skills are modular instruction packages that extend AI assistant capabilities. Each skill provides specialized domain knowledge, workflows, and best practices that can be loaded on-demand.
Browse all 0 skills across 0 tiles in the Skill Catalog.
Skills are plain Markdown, so any ecosystem installer can add the whole collection to an agent configuration:
# Install all skills into your project
npx skills add ./skills --allEach bundled Rust tool can also install (or remove) its own companion skill in detected agents:
pantheon-skill-auditor skill install # or: skill uninstall
pantheon-adr skill install
pantheon-journal skill installThe skill catalog in this README and the docs tiles page is generated by
scripts/catalog:
# Regenerate the catalog
bun run readme:update
# Preview the changes without writing
bun run readme:update:dry-runcrates/skill-auditor/ is a typed, testable Rust binary that implements the 9-dimension rubric (140 pts total) with unit tests and structured JSON output. It bundles the crates/skill-validator-rs/ structural checks, so the two crates form the full audit gate. Both build from the Cargo workspace; no separate install is needed.
# Build once (cached in target/release)
cargo build --release -p pantheon-skill-auditor
# Evaluate a single skill
cargo run -p pantheon-skill-auditor -- evaluate agentic-harness/skill-quality-auditor --json --store
# Batch evaluate
cargo run -p pantheon-skill-auditor -- batch infrastructure/terraform-generator ci-cd/github-actions-generator --store
# Exit 1 if any skill grades below B+
cargo run -p pantheon-skill-auditor -- batch --fail-below B+ agentic-harness/skill-quality-auditorbun run build:skill-auditor is a shortcut for the cargo build above. See crates/skill-auditor/ for source.