一键导入
speckittasks
// Generate an actionable, dependency-ordered tasks.md for the feature based on available design artifacts.
// Generate an actionable, dependency-ordered tasks.md for the feature based on available design artifacts.
Analyze indexed codebases via graph database (neug) and vector index (zvec). Covers call graphs, dependencies, dead code, hotspots, module coupling, architecture reports, semantic search, impact analysis, bug root cause from GitHub issues, class diagrams (UML), and PR review (risk scoring, conflict detection, auto-merge candidates, labeling). Also covers creating, inspecting, and repairing a CodeScope index. Use for: code structure, who calls what, why something changed, similar functions, module boundaries, bug tracing, class relationships, PR risk/conflicts, or any question benefiting from a code knowledge graph. Applies when a `.codegraph` index exists in the workspace, or when the user wants to create one.
Submit a new issue to the project.
Submit a new Pull Request to the project.
Analyze new commits and generate related test cases.
Execute the implementation planning workflow using the plan template to generate design artifacts.
Create or update the feature specification from a natural language feature description.
| name | speckit.tasks |
| description | Generate an actionable, dependency-ordered tasks.md for the feature based on available design artifacts. |
| disable-model-invocation | true |
$ARGUMENTS
You MUST consider the user input before proceeding (if not empty).
Locate Feature:
###-feature-name, e.g., 001-feature-name../specs/[###-feature-name]. The feature id and name must be exact match.Pre-check:
./specs/[###-feature-name], otherwise ERROR "Feature directory not found: ./specs/[###-feature-name]"../specs/[###-feature-name]/spec.md, otherwise ERROR "Spec file not found: ./specs/[###-feature-name]/spec.md"../specs/[###-feature-name]/plan.md, otherwise ERROR "Plan file not found: ./specs/[###-feature-name]/plan.md".Execute task generation workflow:
Generate metadata:
./specs/[###-feature-name]/tasks/.templates/tasks-metadata-template.md to the tasks directory as specs/[###-feature-name]/tasks/metadata.mdGenerate modules:
templates/tasks-module-template.md to the module file as specs/[###-feature-name]/tasks/module_i.md.Report: Generate a summary of the tasks:
Context for task generation: $ARGUMENTS
Each task should be immediately executable - each task must be specific enough that an LLM can complete it without additional context.
CRITICAL: Tasks MUST be organized by module and follow the template templates/tasks-metadata-template.md and templates/tasks-module-template.md.
Tests are OPTIONAL: Only generate test tasks if explicitly requested in the feature specification or if user requests TDD approach.
Metadata is saved in tasks/metadata.md. Metadata must follow the template including:
Modules are saved in tasks/module_i.md (one file per module). Module must follow the template including:
Each module file includes multiple tasks. Task must follow the template including:
The is the three-digit feature number at the beginning of the feature directory name, i.e., 001-feature-name, e.g., F001, F002, F003...
The must restart for each module. For example:
The number of tasks is not fixed. You can add more tasks as needed but must match the spec.md.