- Home
- Skills
- Dceoy
- Speckit Agent Skills
- Speckit Tasks
speckit-tasks_skill
- Shell
11
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill dceoy/speckit-agent-skills --skill speckit-tasks- SKILL.md6.5 KB
Overview
This skill generates an actionable, dependency-ordered tasks.md for a feature using available design artifacts. It produces checklist-formatted tasks organized by user story, a dependency graph, parallel execution examples, and a suggested MVP scope. The output is immediately executable by an LLM or developer.
How this skill works
The skill inspects specs/<feature>/plan.md and specs/<feature>/spec.md and optionally reads data-model.md, contracts/, research.md, and quickstart.md. It extracts tech stack, user stories with priorities, entities, and endpoints, then maps components to stories. It produces tasks.md using the prescribed checklist format, phases (Setup, Foundational, per-story phases, Polish), dependency ordering, and independent test criteria. It validates format rules and reports task counts, parallel opportunities, and the generated file path.
When to use it
- You have a completed implementation plan and need a dependency-ordered task list.
- You want tasks organized by user story so increments are independently testable.
- You need a machine-executable checklist for LLM-driven implementation.
- You want a suggested MVP and parallelization opportunities mapped out.
Best practices
- Ensure specs/<feature>/plan.md and spec.md are present; run speckit-plan if missing.
- Include optional artifacts (data-model.md, contracts/, research.md) to improve task accuracy.
- Keep user stories prioritized in spec.md (P1, P2...) so phases map correctly.
- Request tests explicitly if you want test tasks generated (TDD) — tests are optional otherwise.
- Review generated tasks.md for environment-specific paths and adjust before handing to an executor.
Example use cases
- Generate tasks.md for a new feature where plan.md lists libraries and structure so implementers can start immediately.
- Convert spec.md user stories into independent development phases with exact file paths for each task.
- Produce parallel execution examples to speed up delivery across multiple contributors or LLM agents.
- Create MVP scope (usually User Story 1) and a roadmap for incremental delivery and polishing.
FAQ
Required: specs/<feature>/plan.md and specs/<feature>/spec.md. Optional: data-model.md, contracts/, research.md, quickstart.md.
Will the skill produce tests automatically?
No — test tasks are generated only if requested in the feature specification or if you explicitly ask for a TDD approach.