---
title: "skraft-backlog"
description: "skraft backlog: discover and triage GitHub issues, refine them into ready stories, and review each issue's refinement in a comment, from your editor or a GitHub Agentic Workflow."
canonical: https://agentpluginsdirectory.com/plugins/skraft-backlog
last-updated: 2026-10-12
---

# skraft-backlog
skraft backlog: discover and triage GitHub issues, refine them into ready stories, and review each issue's refinement in a comment, from your editor or a GitHub Agentic Workflow.
- Slug: skraft-backlog
- Publisher: SebastienDegodez
- Repository: https://github.com/SebastienDegodez/skraft-plugin
- Manifest: plugins/skraft-backlog/plugin.json
- Version: 1.15.1
- License: GPL-3.0-or-later
- Category (editorial): other
- Skills: 8 (backlog-review-lenses, discovery-review-criteria, github-issue-search, issue-refinement, issue-triage, planning-review-criteria, refinement-proposal, sprint-planning)
- MCP servers: 0
- Stars: 8
- Repository created: 2026-04-16
- Repository last pushed: 2026-10-11
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/skraft-backlog
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What skraft-backlog does, in the publisher's words

The backlog half of SKRAFT, as a plugin of its own: find the GitHub issues worth working on, classify them, and refine them into stories that clear the Definition of Ready. It installs and runs without the engineering plugin (skraft), and the engineering plugin runs without it.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/SebastienDegodez/skraft-plugin/HEAD/plugins/skraft-backlog/README.md

## Skills

- backlog-review-lenses: Use when a triage report, a sprint proposal, refined stories or a refinement proposal for one issue must pass an independent review before it is handed on: dispatching each review lens with only its own inputs, collecting their results and turning them into a verdict with the bundled script. Not fo…
- discovery-review-criteria: Use when reviewing DISCOVER artefacts (triage reports, sprint proposals) for completeness, prioritization quality, and duplicate handling. Contains gate definitions G1-G6 and scoring rubric for the lenses of the backlog-discoverer review gate.
- github-issue-search: Use when finding GitHub issues to work on, exploring a backlog by labels, milestones or assignees, finding issues related to changed files, or writing a triage label or a refinement comment on an existing issue. Covers issue discovery queries, pagination and ranking, and the write route for each ho…
- issue-refinement: Use when transforming raw issues or feature requests into well-structured user stories with acceptance criteria. Covers user story format, INVEST criteria, acceptance criteria patterns, story splitting techniques, DoR 8-item gate, and 8 antipatterns to detect.
- issue-triage: Use when triaging GitHub issues by assigning labels, priority, effort estimates, or detecting duplicates. Covers triage methodology, label taxonomy, priority frameworks, effort sizing, and sprint proposal construction.
- planning-review-criteria: Use when reviewing DISCUSS artefacts (stories, acceptance criteria, sprint plans) for INVEST quality, planning coherence, and DoR compliance. Contains gate definitions G1-G8 and scoring rubric for the the backlog-planner review gate lenses.
- refinement-proposal: Use when reviewing one GitHub issue and proposing its refinement as a comment: Definition of Ready check, Fibonacci size and capacity estimate, ready or needs-refinement verdict, a proposed story with domain examples and acceptance criteria, what is vague or non-compliant in the current criteria, c…
- sprint-planning: Use when planning sprint content, prioritizing stories within milestones, estimating capacity, or analyzing dependency graphs between stories. Covers MoSCoW prioritization, milestone management, velocity tracking, and dependency resolution for GitHub-based workflows.

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.