sync

Synchronize keel documentation with implementation progress and detect doc-code drift.

Updated Jan 12, 2026
One-click install
npx skills add https://github.com/ab300819/keel-workflow --skill sync-ab300819
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sync
Source: https://github.com/ab300819/keel-workflow/tree/main/skills/sync
Command: npx skills add https://github.com/ab300819/keel-workflow --skill sync-ab300819

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? After development tasks are completed, keel documents (requirements, design, test cases, dev tasks) often fall out of sync with the actual codebase. This Skill keeps documentation aligned with implementation progress, detects drift between docs and code, and archives completed content to control document bloat. ## Core Features & Use Cases - Default Sync Pipeline: Runs trace → audit → check automatically, validating traceability matrix change sources, checking numbering integrity (AC coverage, F task closure, orphan IDs), and generating a deviation report with a health score. - Low-Risk Absorption: Automatically absorbs low-risk deviations (status field updates) while requiring user confirmation for high-risk changes like undocumented interfaces or tests. - Document Archiving: Archives completed requirements, designs, tests, and tasks into archive/ files with cascade support and an index, triggered by line-count or completion thresholds. - PRD Back-Propagation: Reverse-syncs keel requirement changes to the PRD mapping table via --back-propagate-prd. - Use Case: After finishing a sprint's development tasks, run the sync command to update task statuses, verify traceability coverage, get a health score, and archive completed features. ## Quick Start Ask the AI to run /sync after completing development tasks to update keel document statuses and generate a deviation report.

Frequently Asked Questions about sync

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

FAQPage Schema
How do I sync documentation with code progress after development?

Run /sync with no arguments to execute the default trace → audit → check pipeline. It scans the workspace state, compares it against keel documents, auto-absorbs low-risk status updates, and asks for confirmation before writing high-risk changes.

How do I archive completed requirements and tasks in keel docs?

Use /sync --archive with an optional document type (requirements, design, tests, tasks). Archiving triggers when thresholds are met, such as more than 15 completed tasks or documents exceeding line limits, and supports cascading archival of related content.

Does sync scan code for requirement IDs in comments?

No. The trace phase deliberately does not scan code for keel numbering annotations. It validates the traceability matrix's change-source column, which records repository@sha references written by dev-workflow commits, keeping the dependency direction docs-to-code only.

What is the difference between sync and verify in keel?

Sync updates document statuses and absorbs deviations based on implementation progress, while verify checks implementation correctness and alignment. Schema drift diagnosis is delegated to /verify --schema-drift, and sync only merges its results into the health report.

Why does sync refuse to archive some completed features?

Features belonging to a milestone that has not passed manual acceptance verification are not archived, even if marked complete. Archiving them would remove the F-to-milestone mapping needed as evidence during milestone acceptance.