init-harnesslisted

One-shot initialization of a parallel-terminal Claude Code build harness in a project - phase checkboxes as the task queue, pinned model/effort per agent, mechanical git-tag gates, hardened read/gate/log hooks, Obsidian memory vault plus SQLite agent memory, human-gated pivots, and CI drift checks. Use when the user says "/init-harness", "init-harness", "initialize the harness", "set up the build harness", "scaffold this project's harness", "bootstrap the agent harness", or drops in a project-context.yml and asks to initialize. Also use when a project needs multi-terminal agent orchestration with gates, or when an existing harness must be verified/repaired (hooks not firing, guards not blocking, .agents drift). Do NOT use for adding a single agent or skill to an existing project.
M4NUSH7/Niche-Claude-Code · ★ 1 · AI & Automation · score 77

Install: claude install-skill M4NUSH7/Niche-Claude-Code

# init-harness Stands up a complete agentic build harness in one pass: **preflight -> intake -> scaffold -> content -> derive -> verify.** ## The one thing to understand first **This skill is ~80% a config generator.** The heavy lifting already exists and is tested: `sync_harness.py` generates the entire `.agents/` tree, stamps model IDs, and emits the skill indexes and prompt stubs; `harness_status.py --write` generates the Active Context block and `REALITY.md`. The source docs say so outright - *"the config is the only step that requires thought."* So do **not** write a new generator. Write a good `harness.config.json`, run the scripts that exist, and prove the result works. Anything else fails the simplicity ladder at rung two (*already in the codebase?*). ## Division of labour - do not blur it | Layer | Who | What | |---|---|---| | **Structure** | `scripts/scaffold.sh` | ~40 dirs, kit copy, `git init`, sqlite init. **One call, zero judgment.** | | **Content** | **you (the agent)** | config, agent prompts, commands, `CLAUDE.md`/`AGENTS.md`, context packs, `.bld/*/phases.md`, `memory/INDEX.md` | | **Derivation** | existing scripts | `sync_harness.py` -> `.agents/` + indexes + stubs; `harness_status.py --write` | | **Proof** | `scripts/verify_init.py` | guards actually **block** | **Never `mkdir`/copy through tool calls.** Scaffolding directories one-by-one is ~40 round trips of ceremony for zero judgment. Your tokens go to content. (Unless the user explicitly asks fo