From runesmith-sprint
Turn off the Atlassian interconnect for this project - strip applied CLAUDE.md sections, remove markers, remove deployed CC skills. Use when the user says "disable atlassian", "turn off atlassian", "unwire Jira", "remove atlassian from this project".
How this skill is triggered — by the user, by Claude, or both
Slash command
/runesmith-sprint:disableThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Strip the Atlassian-driven workflow from a project. Inverse of `/atlassian:enable`. Idempotent.
Strip the Atlassian-driven workflow from a project. Inverse of /atlassian:enable. Idempotent.
Comms history (archive) is preserved. CC .credentials and any user-edited content outside section markers are preserved.
lib/atlassian-enabled.md - marker semanticslib/jira-apply.md - what gets strippedlib/comms-check.md - runs firstlib/user-prompts.md - structured-input requirement for any user promptThis skill follows the marketplace-wide user-prompt standard in lib/user-prompts.md. Every user prompt MUST use the host client's structured input UI (single-pick, multi-pick, or text-input form). Never freeform plain-text yes/no questions. The only exception is the consent-trigger gate documented in lib/consent.md, which waits for user-initiated phrases like "make the ticket".
If to: user items exist, surface them. User may want to resolve before disabling.
Verify project is currently atlassian-enabled. If not: "Already disabled. Nothing to do."
Disable Atlassian for {PROJECT}?
This will:
- Strip applied sections from <workspace>/CLAUDE.md
- Strip applied sections from {PROJECT}.cc/CLAUDE.md
- Delete <workspace>/.atlassian-enabled
- Set atlassianEnabled: false in {PROJECT}.cc/.claude-code-workspace
- Remove deployed skills: {PROJECT}.cc/.claude/skills/atlassian/
This will NOT:
- Delete {PROJECT}.cc/.credentials (manual cleanup if desired)
- Touch any open or archived comms
- Touch any plans (including their tickets/ subfolder), drafts, or Jira state
Wait for trigger phrase ("disable", "yes", "unwire").
Backup files that will change to archive/_pre-atlassian-disable/<ISO>/.
In each of <workspace>/CLAUDE.md and {PROJECT}.cc/CLAUDE.md:
<!-- atlassian-section:start --> and <!-- atlassian-section:end -->.Read {PROJECT}.cc/.claude-code-workspace. Set:
{
"atlassianEnabled": false,
"atlassian": null
}
Preserve other fields (project, repos, etc.).
.atlassian-enabled filerm <workspace-root>/.atlassian-enabled
Delete {PROJECT}.cc/.claude/skills/atlassian/ directory.
Cowork's Project Instructions UI field is invisible to the agent. Disable cannot edit it directly. It MUST surface a removal instruction for the user.
The Atlassian supplement was wrapped in HTML-comment markers when sprint:enable emitted it. Tell the user to remove everything between (and including) the markers:
─────────────────────────────────────────────────────────────
PROJECT INSTRUCTIONS - atlassian supplement removal
Open Cowork's UI (app sidebar → project settings → Instructions)
and delete the entire block from:
<!-- runesmith:atlassian-start -->
through to and including:
<!-- runesmith:atlassian-end -->
inclusive. The base Project Instructions content stays.
─────────────────────────────────────────────────────────────
✓ Atlassian disabled
Stripped: <workspace>/CLAUDE.md, {PROJECT}.cc/CLAUDE.md
Removed: .atlassian-enabled, .claude/skills/atlassian/
Marker updated: atlassianEnabled: false
Preserved:
{PROJECT}.cc/.credentials (delete manually if desired)
comms/archive/ (audit trail)
plans/, drafts/ (untouched; ticket JSON drafts under plans/active/<slug>/tickets/ untouched)
archive/tickets-pushed/ (ticket history, untouched)
Re-enable any time with /runesmith-sprint:enable.
.credentials not touchedAlready disabled: Exit clean; no-op. Marker file mismatch (workspace says disabled, CC says enabled): Surface the inconsistency, default to disabling both, ask user before continuing. CC head deleted: Workspace marker can still be removed; skip CC operations and warn.
npx claudepluginhub runemarklabs/runesmith --plugin runesmith-sprintGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.