{"id":"cb63236f-e921-4627-97f0-f0a8918a1135","slug":"haulth-agent-orchestration","kind":"skill","name":"agent-orchestration","summary":"","owner":"haulth","repoUrl":"https://github.com/haulth/agent-orchestration","license":"MIT","licenseTier":"permissive","trust":"community","stars":1,"match":null,"fewStars":true,"format":"skill","createdAt":"2026-10-24T11:54:14.000Z","addedAt":"2026-10-26T19:51:22.575Z","logoUrl":"https://avatars.githubusercontent.com/u/105440758?u=5da69841051b476d7c3c026fe471e21978c3a0cf&v=4","linkOnly":false,"collection":null,"skillPath":null,"categories":["writing"],"tags":[],"runtime":"node","local":true,"cloudCapable":false,"cloudCapability":"unknown","cloudReadyOnPlatform":false,"needsApiKey":false,"reviewed":false,"syncedAt":"2026-10-26T19:51:30.373Z","sample":false,"url":"https://conduits.dev/tools/haulth-agent-orchestration","language":"Python","transports":["stdio"],"install":{"npm":"agent-orchestration-skill"},"requirements":{"env":[],"upstreamMayCharge":false},"tools":null,"readme":"# agent-orchestration\n\nMulti-agent and sub-agent orchestration skill for **Claude Code**, **Codex CLI**, and **Antigravity (`agy`)**. Designed for hardware-aware parallel execution, dynamic zero-token model routing, structural work decomposition (**Orchestration Anatomy**), prerequisite/threshold keyhole interlocks, direct parent-to-child skill sharing, and an ephemeral skill sandbox.\n\n## Core Capabilities\n\n- **Zero-Token Live Model Discovery & Routing (`discover.py` + `route.py`):** Automatically probes available models across installed harnesses (`route.py --list-models`) and selects the optimal model per tier (`fast`, `standard`, `frontier`, `frontier_max`), supporting cross-vendor, single-host (`--single-host`), and single-model (`--same-model`) operation with Zero-Prior de-biased briefs.\n- **Orchestration Anatomy & Keyhole Interlocks (`anatomy.py` + `batch --audit`):**\n  - **Upfront Sizing Gate (`Micro` <=1 | `Figure` 2–4 | `Hero` 5–8 | `Epic` >8):** Classifies plan scale up front and gates `Epic` plans (`>8` packages) behind a `Hero` slice proposal unless explicitly approved (`--allow-epic`).\n  - **Truth Lock & Keyhole Invariants (`truth_contract` + `--invariant` + `truth_hash`):** Locks user prerequisites, shared constants, and numeric threshold bounds (`KEY=VAL`, `KEY<=VAL`, `KEY>=VAL`, `must:PAT`, `forbid:PAT`) in `.orch/truth-contract.md` with a SHA-256 `truth_hash` and write-locks the contract against unilateral worker edits.\n  - **Topological Wave Settler & Audit to Zero (`settle` + `audit`):** Resolves explicit and implicit `outputs -> context_file` dependencies into parallel execution waves (`Wave 0, Wave 1, ...`) and proves **0 structural and interlock defects** across 8 checks (`clearance`, `cycles`, `terminals`, `support`, `stand_in_briefs`, `skill_mounts`, `truth_contract`, `interlocks`) before running any agent.\n  - **Anti-Silent-Mutation & Change-Ripple Protocol (`amend` + `inspect`):** Scans deliverables after each wave to block threshold drift (`INVARIANT_DRIFT`, `THRESHOLD_VIOLATION`) and conflicting constants across packages (`CROSS_PACKAGE_CONFLICT`). When a locked parameter must change with empirical proof (`truth_amendments`), `anatomy.py amend` (or `batch --auto-amend`) increments `truth_version`, broadcasts `.orch/truth-amendment-vN.md` into all packages' `context_file`, and flags already-run packages with the old value in `stale_packages` for mandatory resync.\n- **Direct Parent-to-Child Skill Sharing (`--skill`):** Discovers specialized skills already available on the parent agent (`route.py --list-skills`) and mounts their `SKILL.md` path and directory (`--add-dir`) directly into child agents without redundant installation.\n- **Ephemeral Skill Sandbox (`skill_sandbox.py`):** When a specialized skill is not installed locally, searches trusted repositories (`anthropics/skills`, `openai/skills`, `vercel-labs/agent-skills`, `huggingface/skills`, `obra/superpowers`, or verified GitHub repositories with `>= 50` stars), requires explicit user approval (`--approved-by-user`), runs static security checks, and installs into isolated skill memory (`~/.agent-orchestration/sandbox-skills/`) that can be removed or purged at any time.\n- **Hardware-Aware Parallel Execution (`probe_hardware()`):** Dynamically probes CPU cores, available RAM, free disk space, and GPU VRAM to cap parallel `wip` safely, isolating concurrent writes via disjoint `--may-write` scopes or Git worktrees (`--worktree`).\n- **Anti-Hallucination Evidence Law (`No Proof = No Claim`):** Every conclusion and premise challenge (`assumptions_challenged`) must cite verifiable ground truth (`file:line`, `command -> output`, or `DOI/arXiv/PMID/URL`).\n\n## Installation\n\n```bash\n# Option 1: Install via Agent Skills CLI\nnpx skills add haulth/agent-orchestration\n\n# Option 2: Install / inspect directly via NPX (npm registry or GitHub)\nnpx agent-orchestration-skill\nnpx agent-orchestration-skill doctor\n```\n\nOr install from a local clone (syncs to `~/.claude/skills`, `~/.agents/skills`, `~/.codex/skills`, and `~/.gemini/config/skills`):\n\n```powershell\n.\\install\\install.ps1   # Windows PowerShell\n```\n```bash\n./install/install.sh    # macOS / Linux\n```\n\n## Quick Start Commands\n\n```bash\n# 1. Inspect available models, parent skills, and optimal routing (0 tokens)\npython <skill>/scripts/route.py --list-models\npython <skill>/scripts/route.py --list-skills\npython <skill>/scripts/route.py --task-type research --need arxiv --manager agy --same-model\n\n# 2. Orchestration Anatomy: Init Truth Contract + Keyhole Invariants, Audit to Zero, Execute & Inspect 4x Seams\npython <skill>/scripts/anatomy.py init --goal \"Implement & verify distributed service\" \\\n       --invariant \"PORT=8080\" --invariant \"MAX_LATENCY_MS<=200\" \\\n       --pkg \"contract:guided::src/types.py::.orch/contract-out.md\" \\\n       --pkg \"impl:guided:contract:src/core/::src/core/engine.py\" \\\n       --pkg \"verify:exploratory:impl\" --out .orch/plan.json\npython <skill>/scripts/anatomy.py audit .orch/plan.json\npython <skill>/scripts/agent_run.py batch .orch/plan.json [--audit] [--auto-amend]\npython <skill>/scripts/anatomy.py amend .orch/plan.json --set \"MAX_LATENCY_MS<=250\" \\\n       --reason \"Measured p99 cold start\" --evidence \"bench.py:42 -> 235ms\"\npython <skill>/scripts/anatomy.py inspect .orch/plan.json\n\n# 3. Run a single worker with parent skill sharing (--skill) and shared blackboard (--context-file)\npython <skill>/scripts/agent_run.py run --agent agy --brief brief.md --mode read \\\n       --skill literature-search-arxiv --context-file .orch/spec.md\n\n# 4. Ephemeral Skill Sandbox (used only when parent lacks a skill; requires user approval before install)\npython <skill>/scripts/skill_sandbox.py check <keyword>\npython <skill>/scripts/skill_sandbox.py search <query> --min-stars 50\npython <skill>/scripts/skill_sandbox.py install <owner/repo> --subpath <path> --approved-by-user\npython <skill>/scripts/skill_sandbox.py list\npython <skill>/scripts/skill_sandbox.py remove <name>\npython <skill>/scripts/skill_sandbox.py purge [--expired]\n```\n\n## License\n\nMIT\n"}