菜单
Run a request through the full FABLE-HARNESS lifecycle (triage -> profile -> taxonomy -> generate per section -> judge -> verify -> missability -> finalize). Auto-invocable for substantive feature/bugfix/refactor requests; the actual orchestration logic lives in .claude/workflows/run.js, invoked here via the Workflow tool.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | run |
| description | Run a request through the full FABLE-HARNESS lifecycle (triage -> profile -> taxonomy -> generate per section -> judge -> verify -> missability -> finalize). Auto-invocable for substantive feature/bugfix/refactor requests; the actual orchestration logic lives in .claude/workflows/run.js, invoked here via the Workflow tool. |
| model | sonnet |
This skill is a thin wrapper. All real orchestration logic lives in .claude/workflows/run.js (the actual Claude Code Workflow-tool primitive — do not confuse this with .claude/team-configs/*.yaml, which is plain data, not a primitive itself).
run_id using the current date/time (e.g. via Get-Date -Format yyyyMMdd-HHmmss or the Bash date equivalent), suffixed with a short slug from the request. Example: 20260710-183000-add-auth-endpoint..fable/<run_id>/ (and its artifacts/, stages/, verdicts/ subdirectories) if it doesn't already exist.<run_id> (no trailing newline) to .fable/current-run so the Stop-hook verifier gate (stop-verify-gate.ps1) knows which run is active.Workflow({ name: "run", args: { run_id: "<run_id>", request: "<the user's request text>" } }).status, stageResults, missability) to the user in plain language — don't just dump the raw JSON..fable/current-run (or leave it if the user wants to keep verifying/iterating further — use judgment; if genuinely done, clear it so the Stop hook stops gating on a finished run).Per N7, every stage inside run.js dispatches a typed agent — never a generic stand-in. Per N2, nothing in this pipeline ever touches Fable-5; if the request is unusually high-stakes and the user might benefit from /plan-deep instead of (or before) /run, say so as a suggestion, never as an automatic substitution.
Canonical model-tier reference and the Fable-5 manual-only enforcement protocol. Consult before writing or changing any model field (agent frontmatter, skill frontmatter, Task dispatch parameter, settings.json availableModels, env vars). Auto-invocable — any agent or skill authoring a new artifact should load this first.
Templates and risk-tier evolution policy for authoring new agents, skills, and hooks — the 3 artifact kinds this harness produces (no "mcp" kind; this harness has no MCP). Used primarily by meta-agent when extending the harness. Auto-invocable.
Reference-only skill documenting where artifacts live, the 3-verdict vocabulary (pass/pass-with-notes/reject), and the Preserved-Invariants-vs-Changed-Behaviors contract for revising an existing artifact. Load before writing or revising any artifact under .fable/. Auto-invocable.
Show CONSTITUTION.md's content and invariant list on request, re-run the SHA-256 hash check against CONSTITUTION.sha256 on demand, and document the amendment procedure. Useful for Claude to consult proactively as well as for a user to invoke directly.
Merge-safe installer that makes FABLE-HARNESS's agents/skills/hooks available at Claude Code's user scope (~/.claude), so they work across every project, not just H:\FABLE-HARNESS. Manual-only — never auto-invoked, since this touches shared, hard-to-reverse state outside this project.
Highest-stakes planning skill using Fable-5. Manual-only — Claude can never auto-invoke this; only a user typing /plan-deep can. Use for the small number of asks big enough to warrant Fable-level structure and thinking budget (major-scope architecture, cross-cutting rewrites, irreversible decisions).