speckit-companion-implement

Executes spec-kit tasks.md phases in dependency order and marks specs complete.

Updated May 28, 2026
One-click install
npx skills add https://github.com/whw23/PingoGate --skill speckit-companion-implement-whw23
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: speckit-companion-implement
Source: https://github.com/whw23/PingoGate/tree/main/.claude/skills/speckit-companion-implement
Command: npx skills add https://github.com/whw23/PingoGate --skill speckit-companion-implement-whw23

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manually working through a spec-kit tasks.md file is error-prone: tasks get done out of dependency order, checkboxes drift from actual progress, and lifecycle timing data in .spec-context.json gets corrupted by hand edits. This Skill automates the implement phase so execution, logging, and completion stay consistent. ## Core Features & Use Cases - Wave-based task execution: Reads tasks.md phases (Setup, Foundational, user stories by priority, Polish) and executes tasks wave by wave, respecting ⟶ Wait dependency join lines. - Finish-only event logging: Appends each task completion to .spec-context.events.jsonl via write-context.py, then materializes waves to check off tasks.md boxes without hand edits. - Lifecycle completion: Validates work against the spec's Functional Requirements and Success Criteria, then marks the spec complete through the sanctioned --mark-complete writer. - Use Case: After running spec-kit plan and tasks commands for a feature, invoke this Skill to build every task in order, keep the companion panel's timing data accurate, and land the spec at status completed. ## Quick Start Run the speckit companion implement step to execute all tasks in my current feature's tasks.md in dependency order and mark the spec complete.

Frequently Asked Questions about speckit-companion-implement

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

FAQPage Schema
How do I execute a spec-kit tasks.md file automatically?▼

Run the companion implement command, which reads tasks.md and executes each phase in dependency order: Setup, Foundational, user stories by priority, then Polish. Tasks within a wave are independent and a Wait line marks where the next wave depends on completed work.

How does spec-kit track task completion during implementation?▼

Each task finish is appended to .spec-context.events.jsonl via write-context.py with a one-line summary and touched files. A per-wave --materialize call folds those finishes into the panel and checks off the matching tasks.md checkboxes idempotently.

Can I parallelize spec-kit implementation tasks with subagents?▼

Yes, if the host supports subagents and a wave's tasks are substantial enough to justify worker startup. Each subagent makes only its task's edits and appends its own finish event. For small quick edits, inline execution is the default and usually faster.

What happens if python3 is unavailable during spec-kit implement?▼

The mark-complete step is best-effort: if python3 is unavailable it warns and skips without failing the host command. Hook checks and node hooks also degrade gracefully and never abort the pipeline.

Why should I not hand-edit tasks.md checkboxes or .spec-context.json?▼

Hand edits corrupt shared state, such as duplicated status keys in .spec-context.json. The write-context.py script owns both files: it stamps real clock times, writes atomically, and flips checkboxes from journaled finishes so the journal and tasks.md never diverge.