Arete
Official@Arete-Consortium
Arete publishes 144 Claude Code skills spanning multi-agent orchestration, release engineering, security auditing, DevOps, web development, and niche business-domain personas.
Agent Skills by Arete
Showing 141 vetted skills indexed across 1 GitHub repositories.
session-start
Audits project context, git state, and pending work to bootstrap engineering sessions.
agent-pipeline
Automates Issue-to-PR-to-Merge pipelines using GitHub Actions and AI agents with review loops.
release-engineering
Orchestrates preflight checks, changelog generation, version bumps, tagging, and publishing for software releases.
session-end
Wraps up Claude Code sessions by logging decisions, capturing learnings, and committing work.
exploration
Generates distinct decision options with evidence, unknowns, and assumption challenges.
specification
Generates fixed-structure technical specifications with measurable acceptance criteria from gathered inputs.
feature-implementation
Orchestrates end-to-end feature implementation through a six-phase WHY/WHAT/HOW workflow with quality gates.
context-mapper
Maps codebase structure, dependencies, and conventions into a YAML context map before code changes.
production
Generates final deliverables from structured inputs with no preamble, filler, or meta-commentary.
feature-pipeline
Orchestrates multi-phase feature development with checkpoint gates and parallel agent streams.
evaluation
Scores candidates against a weighted rubric and produces an evidence-cited decision verdict.
conductor
Orchestrates context-driven development with spec generation, TDD implementation, and checkpointed reverts.
goal
Manages a durable build objective with budget and turn ceilings via a repo-root sentinel file.
release-engineer
Automates release readiness checks, changelog generation, version tagging, and registry publishing for git repositories.
document-forensics
Analyze document collections for provenance, anomalies, redactions, and cross-document contradictions.
technical-debt-auditor
Audits repositories for technical debt and generates scored DEBT.md reports with ROI-ordered fixes.
handoff
Packages project state into structured handoff documents for agent sessions and human pickup.
entity-resolver
Consolidates duplicate entity mentions across document corpora using fuzzy matching and confidence scoring.
intent-author
Authors and validates machine-comparable intent nodes for Convergent's intent graph.
ci-diagnostician
Diagnose CI build failures and propose minimal fixes from GitHub Actions and GitLab logs.
workflow-debugger
Diagnoses failed Gorgon workflow runs by analyzing checkpoint state, agent logs, and budget traces.
github-operations
Manage GitHub repositories through Git CLI operations with branch protection and commit conventions.
api-client
Makes authenticated HTTP requests to REST APIs with retry logic, rate limiting, and pagination.
process-runner
Execute shell commands with blocklist validation, timeouts, and structured output capture.
Frequently Asked Questions About Arete
FAQPage SchemaWhat tasks can I accomplish with Arete's skill collection?▼
Arete's 144 skills cover session management, feature implementation pipelines, code review, release engineering, CI failure diagnosis, security auditing, web scraping, database operations, OAuth integration, web design and deployment, SEO content pipelines, data analysis, and niche domains like EVE Online integration and hauling-business operations.
Who is the target audience for Arete's skills?▼
The skills target software engineers, DevOps practitioners, security reviewers, and web developers working in Claude Code. Personas include senior software engineers, release engineers, security auditors, CI diagnosticians, web designers, data analysts, and domain coaches for small businesses like junk removal and tie-dye.
How do I install and run Arete's skills?▼
Skills are invoked as slash commands or personas inside Claude Code sessions, such as /release, /security-auditor, /test, or /deploy. Most are user-invocable with cross-platform support on macOS, Linux, and Windows, and many run autonomously or under supervised trust levels with defined risk tiers.
What prerequisites and dependencies do Arete's skills require?▼
Skills run within Claude Code and rely on standard tool permissions like Read, Write, Edit, Bash, Glob, Grep, WebFetch, and WebSearch. Specific skills depend on external scanners (semgrep, bandit, gitleaks, pip-audit), Git CLI, GitHub Actions, pytest, ruff, or platform SDKs such as Sui for EVE Frontier on-chain work.
Are Arete's skills stable and safe for production use?▼
Most skills are marked experimental, with a subset (session-start, security-auditor, code-reviewer, software-architect, testing-specialist) marked stable. Each skill declares a risk level and trust mode — autonomous, supervised, or consensus-gated — and high-risk actions like email sending require unanimous-plus-user approval.