Xerxes-Agents

Add a Bun-native YAML agent definition with inheritance, tool policy, and subagent references.

Add a Bun-native messaging channel adapter with lifecycle, webhook, and configuration coverage.

Add a Bun-native LLM provider entry with routing, pricing, headers, limits, and Bun tests.

Add a Bun-native TypeScript tool with schema, registry wiring, policy boundaries, and Bun tests.

Author a Bun-native Xerxes SKILL.md bundle with valid metadata, safe assets, and discovery tests.

Normalize Xerxes Apache-2.0 headers with the native Bun maintenance command and verify TypeScript sources.

Manage Apple Notes via the memo CLI on macOS (create, view, search, edit).

Manage Apple Reminders via remindctl CLI (list, add, complete, delete).

Generate professional dark-themed system architecture diagrams as standalone HTML/SVG files. Self-contained output with no external dependencies. Based on Cocoon AI's architecture-diagram-generator (MIT).

Search arXiv papers by query, author, category, or identifier using the native Bun client.

Generate ASCII art using pyfiglet (571 fonts), cowsay, boxes, toilet, image-to-ascii, remote APIs (asciified, ascii.co.uk), and LLM fallback. No API keys required.

Production pipeline for ASCII art video — any format. Converts video/audio/images/generative input into colored ASCII character video output (MP4, GIF, image sequence). Covers: video-to-ASCII conversion, audio-reactive music visualizers, generative ASCII art animations, hybrid video+audio reactive, text/lyrics overlays, real-time terminal rendering. Use when users request: ASCII video, text art video, terminal-style video, character art animation, retro text visualization, audio visualizer in ASCII, converting video to ASCII art, matrix-style effects, or any animated ASCII output.

Use when user types /autoresearch, /autoresearch:plan, /autoresearch:debug, /autoresearch:fix, /autoresearch:security, /autoresearch:ship, /autoresearch:scenario, /autoresearch:predict, /autoresearch:learn, or /autoresearch:reason, or mentions "autoresearch" with a goal/metric. Autonomous Goal-directed Iteration — apply Karpathy's autoresearch principles to ANY task: modify, verify, keep/discard, repeat. Supports bounded mode via Iterations: N inline config.

Autonomous multi-iteration bug-hunting and repair campaign for large codebases. Use when the user asks to find bugs, fix regressions, audit an entire project, run many iterations such as 100 rounds, spawn large swarms of agents, create bug bounty reports, or repeatedly scan, patch, verify, and continue until clean or blocked.

Delegate coding tasks to Claude Code (Anthropic's CLI agent). Use for building features, refactoring, PR reviews, and iterative coding. Requires the claude CLI installed.

Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.

Delegate coding tasks to OpenAI Codex CLI agent. Use for building features, refactoring, PR reviews, and batch issue fixing. Requires the codex CLI and a git repository.

Spawns a swarm of specialized agents to deeply analyze the current project and compile comprehensive findings into project-scoped agent memory. ALWAYS runs agents in parallel — sequential analysis is unacceptable.

Swarm-orchestration workflow for Xerxes when a task is too large, ambiguous, long-running, or important for one linear agent. Use when the user asks for autonomous execution, deep implementation, broad project repair, large-codebase work, parallel research, implementation, review, testing, or when the agent should act as the owner/coder and spawn as many subagents as needed while preserving memory, verification, and integration.

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform

Upload an Excalidraw JSON document and return a shareable link without a Python dependency.

Find nearby public places through OpenStreetMap using the native Bun client.

Track Apple devices and AirTags via FindMy.app on macOS using AppleScript and screen capture.

Inspect and analyze codebases using pygount for LOC counting, language breakdown, and code-vs-comment ratios. Use when asked to check lines of code, repo size, language composition, or codebase stats.

Use GitHub authentication that the user or host has explicitly supplied through gh, an injected host integration, SSH, or the current process environment.

Review code changes by analyzing git diffs, leaving inline comments on PRs, and performing thorough pre-push review. Works with gh CLI or falls back to git + GitHub REST API via curl.

Create, manage, triage, and close GitHub issues. Search existing issues, add labels, assign people, and link to PRs. Works with gh CLI or falls back to git + GitHub REST API via curl.

Full pull request lifecycle — create branches, commit changes, open PRs, monitor CI status, auto-fix failures, and merge. Works with gh CLI or falls back to git + GitHub REST API via curl.

Clone, create, fork, configure, and manage GitHub repositories. Manage remotes, secrets, releases, and workflows. Works with gh CLI or falls back to git + GitHub REST API via curl.

Use Gmail, Calendar, Drive, Contacts, Sheets, and Docs through the native Bun Google Workspace client.

Plan and orchestrate GRPO training through a native Bun workflow with explicit accelerator ownership.

Send and receive iMessages/SMS via the imsg CLI on macOS.

GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements.

Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and build multi-step workflows with Guidance - Microsoft Research's constrained generation framework

Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10x speedup vs PyTorch on CPU.

Remove refusal behaviors from open-weight LLMs using OBLITERATUS — mechanistic interpretability techniques (diff-in-means, SVD, whitened SVD, LEACE, SAE decomposition, etc.) to excise guardrails while preserving reasoning. 9 CLI methods, 28 analysis modules, 116 model presets across 5 compute tiers, tournament evaluation, and telemetry-driven recommendations. Use when a user wants to uncensor, abliterate, or remove refusal from an LLM.

Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers, vLLM), and maximize inference speed with Outlines - dottxt.ai's structured generation library

Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.

Use a live Jupyter kernel for stateful, iterative Python execution via hamelnb. Load this skill when the task involves exploration, iteration, or inspecting intermediate results — data science, ML experimentation, API exploration, or building up complex code step-by-step. Uses terminal to run CLI commands against a live Jupyter kernel. No new tools required.