AGENT SKILL

ln-300-task-coordinator

Orchestrates task operations. Analyzes Story, builds optimal plan (1-8 implementation tasks), delegates to ln-301-task-creator (CREATE/ADD) or ln-302-task-replanner (REPLAN). Auto-discovers team ID.

License 许可
MIT
最近更新
2026-05-02
Stars
★ 3
Forks
1

01 前言 Front Matter

name
ln-300-task-coordinator
license
MIT
description
Orchestrates task operations. Analyzes Story, builds optimal plan (1-8 implementation tasks), delegates to ln-301-task-creator (CREATE/ADD) or ln-302-task-replanner (REPLAN). Auto-discovers team ID.

02 资源文件 5

03 SKILL.md 正文

Paths: File paths (shared/, references/, ../ln-*) are relative to skills repo root. If not found at CWD, locate this SKILL.md directory and go up one level for repo root.

Linear Task Planner (Orchestrator)

Coordinates creation or replanning of implementation tasks for a Story. Builds the ideal plan first, then routes to workers.

Inputs

Input Required Source Description
storyId Yes args, git branch, kanban, user Story to process

Resolution: Story Resolution Chain. Status filter: Backlog, Todo

Purpose & Scope

  • Auto-discover Team ID, load Story context (AC, Technical Notes, Context)
  • Build optimal implementation task plan (1-8 implementation tasks; NO test/refactoring tasks) in Foundation-First order
  • Detect mode and delegate: CREATE/ADD -> ln-301-task-creator, REPLAN -> ln-302-task-replanner
  • Strip any Non-Functional Requirements; only functional scope becomes tasks
  • Never creates/updates Linear or kanban directly (workers do)

Task Storage Mode

MANDATORY READ: Load shared/references/tools_config_guide.md, shared/references/storage_mode_detection.md, and shared/references/input_resolution_pattern.md

Extract: task_provider = Task Management → Provider (linear | file).

Workers (ln-301, ln-302) handle the actual Linear/File operations based on task_provider.

When to Use

  • Need tasks for a Story with clear AC/Technical Notes
  • Story requirements changed and existing tasks must be updated
  • Only for implementation tasks (test tasks → ln-404, refactoring → quality gate)

Quality Criteria

MANDATORY READ: Load shared/references/creation_quality_checklist.md §Task Creation Checklist for validation criteria that ln-310 will enforce.

Workflow (concise)

  • Phase 1 Discovery: Auto-discover Team ID (docs/tasks/kanban_board.md). Resolve storyId: Run Story Resolution Chain per guide (status filter: [Backlog, Todo]).
  • Phase 2 Decompose (always): MANDATORY READ: shared/references/goal_articulation_gate.md — Before building IDEAL plan, state REAL GOAL of this Story in one sentence (the deliverable, not the process). Verify: does the decomposition serve THIS goal? Then: Load Story (AC, Technical Notes, Context), assess complexity, build IDEAL plan (1-8 implementation tasks only), scan for reusable patterns (Grep src/ for error handlers, validators, utilities relevant to task categories — count only; if found, append **Pattern Hint:** {count} existing {category} patterns in src/. Review for reuse before creating new (Step 4a in ln-401). to relevant task descriptions), apply Foundation-First execution order, validate Task Independence, assign Parallel Groups, define verification methods for each task AC, extract guide links.
  • Phase 3 Check & Detect Mode: Query Linear for existing tasks (metadata only). Detect mode by count + user keywords (add/replan).
  • Phase 4 Delegate: Call the right worker with Story data, IDEAL plan/append request, guide links, existing task IDs if any; autoApprove=true.
  • Phase 5 Verify: Ensure worker returns URLs/summary and updated kanban_board.md; report result.

Task Plan Readiness Score

Context: Validates plan quality before delegation to workers, preventing rework.

After building IDEAL plan (Phase 2), score 7 criteria:

# Criterion Check
1 Independence No forward dependencies between tasks (Task N uses only 1..N-1)
2 AC clarity Each task AC has measurable outcome AND verification method (test/command/inspect)
3 Tech confidence All referenced technologies/patterns are known or researched
4 Scope isolation Tasks don't overlap with sibling Stories' tasks. Load siblings (list_issues project=Epic.id), compare Affected Components and file paths for structural overlap
5 Architecture compliance Tasks reference correct layers (DB→Repo→Service→API), no planned cross-layer violations (e.g., API task doing direct DB calls)
6 Parallel groups valid Tasks in same group have no mutual dependencies; all deps point to earlier groups; numbers sequential
7 Destructive op safety Tasks with data deletion/migration/schema changes include safety plan (backup, rollback, blast radius)

Score = count of PASS criteria (0-7)

  • 6-7/7: Delegate to worker
  • 4-5/7: Show warnings to user, fix or proceed
  • <4/7: Rework plan before delegation

Verification Methods for Task AC

Context: Goal-Driven Execution pattern — define HOW to verify each AC at planning time so executor (ln-401) can loop through verifications after implementation.

When building IDEAL plan (Phase 2), each task AC must include a verify: method:

Method When to Use Example
test Existing test covers AC verify: test (test_auth.py::test_login_success)
command CLI command validates outcome verify: command (curl -X POST /users → 201)
inspect File/output check verify: inspect (migration file has email column)

Rule: At least 1 AC per task must use test or command (not all inspect).

MANDATORY READ: Load shared/references/ac_validation_rules.md §5 for full format and examples.

Task Independence Validation

Rules per creation_quality_checklist.md #19 (dependencies) and #13 (Foundation-First order).

Examples:

  • ❌ WRONG: "Task 2: Validate token (requires Task 3 to generate keys)"
  • ✅ RIGHT: "Task 1: Generate keys" → "Task 2: Validate token (uses Task 1 keys)"

If forward dependency detected: Reorder, refactor to remove dependency, or split into sequential parts.

Parallel Group Assignment

After building IDEAL plan and validating independence, assign Parallel Group numbers to enable concurrent execution in ln-400.

Algorithm:

group = 1
FOR EACH task T IN ordered_plan:
  deps = tasks that T depends on (from Related/Context)
  IF any dep is in CURRENT group:
    group++
  T.parallel_group = group

Example:

Task Dependencies Group
T1: DB migration none 1
T2: UserRepo T1 2
T3: ProductRepo T1 2
T4: UserService T2 3
T5: API endpoint T4 4

Rules:

  • Tasks in the same group have NO mutual dependencies (only depend on previous groups)
  • Group numbers are sequential (1, 2, 3...), no gaps
  • Single-task groups are valid (sequential execution, same as current behavior)
  • Write **Parallel Group:** {N} in each task document (per shared/templates/task_template_implementation.md)
  • Backward compatibility: if task lacks **Parallel Group:** field, ln-400 treats it as its own group (sequential)

Mode Matrix

Condition Mode Delegate Payload
Count = 0 CREATE ln-301-task-creator taskType=implementation, Story data, IDEAL plan, guideLinks
Count > 0 AND "add"/"append" ADD ln-301-task-creator taskType=implementation, appendMode=true, newTaskDescription, guideLinks
Count > 0 AND replan keywords REPLAN ln-302-task-replanner taskType=implementation, Story data, IDEAL plan, guideLinks, existingTaskIds
Count > 0 AND ambiguous ASK Clarify with user —

Plan Mode Behavior

When invoked in Plan Mode (read-only):

  • Execute Phases 1-3 normally (Discovery, Decompose, Check Existing)
  • Phase 4: DO NOT delegate to workers — instead show IDEAL plan preview:
    • Task titles, goals, estimates, Foundation-First order
    • Mode detected (CREATE/ADD/REPLAN)
    • What worker WOULD be invoked (ln-301 or ln-302)
  • Phase 5: Write plan summary to plan file (not Linear)
  • NO Linear API calls, NO kanban updates, NO worker invocations

TodoWrite format (mandatory): Add phases to todos before starting:

- Phase 1: Discovery (in_progress)
- Phase 2: Decompose & Build IDEAL Plan (pending)
- Phase 3: Check Existing & Detect Mode (pending)
- Phase 4: Delegate to ln-301/ln-302 (pending)
- Phase 5: Verify worker result (pending)

Mark each as in_progress when starting, completed when done.

Critical Rules

  • Decompose-first: always build IDEAL plan before looking at existing tasks.
  • Foundation-First execution order per creation_quality_checklist.md #13.
  • Task limits: 1-8 implementation tasks, 3-5h each (3-5 tasks optimal). Test task created later by test planner.
  • Linear creation must be sequential: create one task, confirm success, then create the next (no bulk) to catch errors early.
  • HARD CONSTRAINT: This skill creates ONLY implementation tasks (taskType=implementation). NEVER include test tasks, manual testing tasks, or refactoring tasks in the plan. Test tasks are created LATER by test planner (after manual testing passes). Refactoring tasks are created by quality gate when code quality issues found.
  • No code snippets in descriptions; workers own task documents and Linear/kanban updates.
  • Language preservation: keep Story language (EN/RU) in any generated content by workers.

Definition of Done (orchestrator)

  • Team ID discovered; storyId resolved (per input_resolution_pattern.md).
  • Story loaded; IDEAL plan built (1-8 implementation tasks only) with Foundation-First order and guide links.
  • NO test or refactoring tasks in IDEAL plan (only taskType=implementation).
  • Existing tasks counted; mode selected (CREATE/ADD/REPLAN or ask).
  • Worker invoked with correct payload and autoApprove=true.
  • Worker summary received (Linear URLs/operations) and kanban update confirmed.
  • Next steps returned (ln-310-multi-agent-validator, then orchestrator continues).

Meta-Analysis

MANDATORY READ: Load shared/references/meta_analysis_protocol.md

Skill type: planning-coordinator. Run after all phases complete. Output to chat using the planning-coordinator format.

Reference Files

  • Tools config: shared/references/tools_config_guide.md
  • Storage mode operations: shared/references/storage_mode_detection.md
  • [MANDATORY] Problem-solving approach: shared/references/problem_solving.md
  • Orchestrator lifecycle: shared/references/orchestrator_pattern.md
  • Auto-discovery patterns: shared/references/auto_discovery_pattern.md
  • Decompose-first pattern: shared/references/decompose_first_pattern.md
  • Plan mode behavior: shared/references/plan_mode_pattern.md
  • Numbering conventions: shared/references/numbering_conventions.md (Task per-Story numbering)
  • Templates (centralized): shared/templates/task_template_implementation.md
  • Local copies: docs/templates/task_template_implementation.md (in target project, created by workers)
  • Replan algorithm details: ln-302-task-replanner/references/replan_algorithm.md
  • Auto-discovery notes: CLAUDE.md, docs/tasks/kanban_board.md

Version: 4.0.0 Last Updated: 2026-02-03

04 同仓库更多 Skills

levalencia
levalencia/agent-god-mode

zarr-python

Chunked N-D arrays for cloud storage. Compressed arrays, parallel I/O, S3/GCS integration, NumPy/Dask/Xarray compatible, for large-scale scientific computing pipelines.

★ 3
levalencia
levalencia/agent-god-mode

zabbix

Zabbix monitoring system automation via API and Python. Use when: (1) Managing hosts, templates, items, triggers, or host groups, (2) Automating monitoring configuration, (3) Sending data via Zabbix trapper/sender, (4) Querying historical data or events, (5) Bulk operations on Zabbix objects, (6) Maintenance window management, (7) User/permission management

开发工程师后端与接口Skill与智能体开发
★ 3
levalencia
levalencia/agent-god-mode

x-twitter-growth

X/Twitter growth engine for building audience, crafting viral content, and analyzing engagement. Use when the user wants to grow on X/Twitter, write tweets or threads, analyze their X profile, research competitors on X, plan a posting strategy, or optimize engagement. Complements social-content (generic multi-platform) with X-specific depth: algorithm mechanics, thread engineering, reply strategy, profile optimization, and competitive intelligence via web search.

★ 3
levalencia
levalencia/agent-god-mode

what-if-oracle

Run structured What-If scenario analysis with multi-branch possibility exploration. Use this skill when the user asks speculative questions like "what if...", "what would happen if...", "what are the possibilities", "explore scenarios", "scenario analysis", "possibility space", "what could go wrong", "best case / worst case", "risk analysis", "contingency planning", "strategic options", or any question about uncertain futures. Also trigger when the user faces a fork-in-the-road decision, wants to stress-test an idea, or needs to think through consequences before committing.

★ 3
levalencia
levalencia/agent-god-mode

vercel-react-best-practices

React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.

前端开发
★ 3
levalencia
levalencia/agent-god-mode

venue-templates

Access comprehensive LaTeX templates, formatting requirements, and submission guidelines for major scientific publication venues (Nature, Science, PLOS, IEEE, ACM), academic conferences (NeurIPS, ICML, CVPR, CHI), research posters, and grant proposals (NSF, NIH, DOE, DARPA). This skill should be used when preparing manuscripts for journal submission, conference papers, research posters, or grant proposals and need venue-specific formatting requirements and templates.

内容创作者写作与文案资料检索与研究
★ 3