init-harnesslisted

One-shot initialization of a parallel-terminal Claude Code build harness in a project - a git worktree and branch per terminal, phase checkboxes as the task queue, pinned model/effort per agent, mechanical git-tag gates, hardened read/gate/log hooks, and a SQLite agent-memory shard per terminal. 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 or 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 72

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

# init-harness **Edition:** CLI (Claude Code). `scripts/*.sh` and `templates/kit/harness/{create,land}-agent-branch` are POSIX shell (macOS / Linux / Git Bash on Windows); every Python entry point is invoked as `<python>`, the verified absolute interpreter from `toolchain.python`, never a bare `python` and never `py -3`. There is no Cowork/Linux-sandbox edition: this skill scaffolds a project's own `.claude/` control plane, which a sandbox session does not own. Seven steps: **preflight -> context -> intake -> scaffold -> content -> derive -> verify.** ## The one thing to understand first **This skill is ~80% a config generator.** Do not write a new generator: write a good `harness.config.json`, run the scripts that exist, prove the result blocks. And never `mkdir`/copy through tool calls - scaffolding ~40 directories one at a time is ~40 round trips of ceremony for zero judgment. Your tokens go to content. | Layer | Who | What | |---|---|---| | Structure | `scripts/scaffold.sh` | dirs, kit copy, `git init` + root commit, `.worktrees/tN`, merge driver, sqlite shards | | Content | **you** | config, agent prompts, commands, CLAUDE.md/AGENTS.md, context packs, `.bld/*/phases.md`, `memory/INDEX.md` | | Derivation | `sync_harness.py`, `harness_status.py --write` | `.agents/`, stamped frontmatter, indexes, status | | Proof | `scripts/verify_init.py` | guards actually **deny** | `token-efficiency` is mandatory here and auto-loads in generated projects: RTK for command output, g