tongtian-lu

Generate a structured talisman execution plan with topology and conflict checks.

10|1|Updated May 4, 2026
One-click install
npx skills add https://github.com/isLinXu/under-one --skill tongtian-lu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tongtian-lu
Source: https://github.com/isLinXu/under-one/tree/main/underone/skills/tongtian-lu
Command: npx skills add https://github.com/isLinXu/under-one --skill tongtian-lu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

It prevents LLM agent workflows from becoming disorganized by turning a vague goal into a structured talisman sequence with an explicit execution topology and risk/cursing constraints.

Core Features & Use Cases

  • Structured task-to-talisman decomposition: Converts a task description or a structured JSON spec into analysis/creation/verification/transformation/retrieval/decision talismans.
  • Mode-based orchestration: Supports orchestration modes like quick-cast, balanced-array, and full-ritual to control coverage, verification tail, and parallelism preference.
  • Conflict, risk, and topology generation: Detects style/dataflow/constraint/SLA conflicts, estimates curse level, and outputs a runnable plan artifact (fu_plan.json).
  • Use case: For a “competitor weekly report” pipeline, generate a deterministic execution blueprint that retrieves inputs, analyzes differences, ranks priorities, drafts the report, validates it, then transforms it to the target format.

Quick Start

Use tongtian-lu to generate a fu_plan.json by providing a task description or task_spec.json, then execute the produced talisman topology.

Frequently Asked Questions about tongtian-lu

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a structured execution plan from a task description for workflow automation?

To generate a structured execution plan for workflow automation, you provide a task description or task specification to decompose the intent into analysis, creation, verification, transformation, retrieval, and decision units, producing a runnable topology plan.

How does conflict detection work across style and dataflow in task orchestration?

Conflict detection in task orchestration works by scanning the generated execution topology to identify style, dataflow, constraint, and performance conflicts across sequential and parallelizable steps before outputting the final plan.

What is the best way to control parallelism and coverage limits in multi-stage planning?

The best way to control parallelism and coverage limits in multi-stage planning is by selecting orchestration modes like quick-cast, balanced-array, or full-ritual, which dictate verification tail depth and mode-driven talisman limits.

Can I use a JSON task specification to create a deterministic report generation pipeline?

Yes, you can use a structured JSON task specification to create a deterministic report generation pipeline by mapping the inputs into a defined sequence of retrieval, analysis, drafting, validation, and transformation steps.

How do risk budget checks prevent workflow assembly errors in agent-ops scenarios?

Risk budget checks prevent workflow assembly errors in agent-ops scenarios by estimating curse levels and validating constraints against the execution topology, ensuring safe assembly across sequential and parallelizable steps.

What are the limitations of using mode-driven limits on talismans for safe workflow assembly?

The limitation of using mode-driven limits on talismans is that orchestration modes restrict the number of execution units and verification depth, which may cap complexity for highly intricate agent-ops pipelines requiring extensive parallelization.