From cortex-core
Manage project backlog items as individual markdown files with YAML frontmatter. Use when user says "/cortex-core:backlog", "backlog add", "backlog list", "backlog pick", "add to backlog", "show backlog", "archive backlog item", "what's ready", "pick a backlog item", or asks to create/view/manage/select backlog items.
How this skill is triggered — by the user, by Claude, or both
Slash command
/cortex-core:backlog <subcommand> [args]<subcommand> [args]The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Manage a file-based project backlog. Each item is a standalone markdown file with YAML frontmatter in `backlog/` (active) or `backlog/archive/` (complete/abandoned).
Manage a file-based project backlog. Each item is a standalone markdown file with YAML frontmatter in backlog/ (active) or backlog/archive/ (complete/abandoned).
Subcommand: $ARGUMENTS (first word = subcommand, remainder = subcommand args; empty = list backlog).
/cortex-core:backlog {{subcommand}} — run the specified subcommand. When {{subcommand}} is add, provide {{title}} to name the new item. When {{subcommand}} is archive, provide {{item}} to identify the target.
Read ${CLAUDE_SKILL_DIR}/references/schema.md when creating or validating items — it contains the field table, enum values, and item template.
Files are named NNN-slug.md where NNN is a zero-padded three-digit sequential ID and slug is a lowercase-kebab-case summary. The numeric prefix is the stable ID used in cross-references (blocks: [7]). Slug drift after title changes is cosmetic and does not break references.
Examples: 001-complete-phase-commits.md, 014-add-search-feature.md
When invoked without a {{subcommand}} (just /cortex-core:backlog), present the available actions via AskUserQuestion:
Create a new backlog item from {{title}}.
cortex-create-backlog-item --title "{{title}}" --status backlog --type {{type}} --priority {{priority}} (type defaults to feature, priority defaults to medium unless user specifies). Pass --parent NNN if the user specified a parent epic.Present the current backlog summary.
backlog/index.mdbacklog/index.md does not exist, suggest running reindex firstUpdate the item's status in place using update_item.py:
cortex-update-item {{item}} status=complete # mark complete
cortex-update-item {{item}} status=abandoned # mark abandoned
This updates frontmatter in place (no file move), cascades blocked-by cleanup, auto-closes parent epics when all children are terminal, and regenerates the index.
Warn the user before closing a backlog or in_progress item — the script will close it
regardless of current status.
Interactive item selector. Presents open backlog items as a selectable list.
cortex-backlog-ready. If exit code is non-zero, parse the error JSON and report the message — suggest running /cortex-core:backlog reindex if the error indicates a missing or malformed backlog index.groups in order (critical → contingent); within each group, iterate items. The first non-empty group's items form the selection set. (Group ordering preserves critical → low priority; within a group, refined items come first.)AskUserQuestion with one question:
label is "NNN — Title" (ID and title)description includes priority and type from the indexAskUserQuestion and note how many additional items were omittedAskUserQuestion:
/cortex-core:lifecycle {{item}} to begin structured developmentin_progress and updated dateReport which items are ready to work on.
cortex-backlog-ready. If exit code is non-zero, parse the error JSON and report the message — suggest running /cortex-core:backlog reindex if the error indicates a missing or malformed backlog index.groups, render a markdown subsection: ### {Priority Title} heading (e.g. ### Critical, ### High, ### Medium, ### Low, ### Contingent) followed by - **{id}** {title} bullets, in iteration order. If all groups are empty, report Backlog is clear.Regenerate the backlog index.
cortex-generate-backlog-indexindex.json and index.md)npx claudepluginhub charleshall888/cortex-command --plugin cortex-coreGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.