AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Feature From Backlog

skill-imtaegan-claude-saas-flow-feature-from-backlog · by ImTaegan

Start a new feature from Linear backlog. Refines backlog item, breaks into sub-tasks if needed, and starts work.

No reviews yet
0 installs
7 views
0.0% view→install

Install

$ agentstack add skill-imtaegan-claude-saas-flow-feature-from-backlog

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-imtaegan-claude-saas-flow-feature-from-backlog)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
1mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Feature From Backlog? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Feature from Backlog — Linear Workflow

Refines backlog items into actionable features and starts work on them.

Project Configuration

Read project configuration from .claude/project.json:

  • linear.teamId, linear.teamName, linear.projectId

If .claude/project.json doesn't exist, suggest running /setup-project first.

Instructions for Claude

Step 0: Load Configuration

PROJECT_ROOT=$(git rev-parse --show-toplevel 2>/dev/null || pwd)

Read $PROJECT_ROOT/.claude/project.json.

Step 1: Show Backlog Items

Use mcp__linear__list_issues:
- team: [from config.linear.teamName]
- project: [from config.linear.projectId]
- state: "Backlog"
- includeArchived: false
- orderBy: "updatedAt"
- limit: 20

Display:

## Backlog Items

Select an item to refine and start working on:

1. **ABC-XX**: [title]
   Priority: [priority or "Not set"]
   Size: [size or "Not estimated"]
   Description: [first line]

[... show all backlog items]

Which item would you like to work on? (Enter number or issue ID)

Step 2: Refine the Selected Item

Fetch full details:

Use mcp__linear__get_issue with selected ID

IMPORTANT: Read and display the FULL issue description. Do NOT truncate or summarize.

Then ask clarifying questions:

Let's refine [Issue ID]: [Issue Title]

1. **What type of work is this?** (new feature, enhancement, integration)
2. **Who will use this?** (admins, users, API consumers)
3. **Is this large enough to need sub-tasks?** (yes/no)
4. **Any specific requirements or constraints?**
5. **Priority level?** (Critical/High/Medium/Low)
6. **Estimated size?** (Small: 1-2h / Medium: 0.5-1d / Large: 2+d)

Step 3: Decide on Approach

If simple feature (Small/Medium, no sub-tasks): skip plan mode, go to Step 4.

If complex feature (Large, needs sub-tasks): do codebase research, then enter plan mode (Step 3a).

Step 3a: Plan Mode for Complex Features

  1. Before planning: Use the Task tool with subagent_type="Explore" and thoroughness="medium" to understand existing patterns and find similar features. Pass the refined feature description as context.
  1. Enter plan mode using EnterPlanMode.
  1. Write the plan file with this EXACT structure:

```markdown # Plan: [Issue ID] — [Issue Title]

## Phase 0: Setup (execute IMMEDIATELY after plan approval)

> These steps MUST be completed before writing any code.

### 0.1 Update parent issue in Linear

  • Tool: mcp__linear__update_issue
  • id: [issue-id]
  • description: [full refined description with requirements and acceptance criteria]
  • labels: ["Feature", priority label, size label]
  • state: "Todo"
  • priority: [0-4]

### 0.2 Create sub-issues in Linear For each implementation step below:

  • Tool: mcp__linear__create_issue
  • team: [config.linear.teamName]
  • project: [config.linear.projectId]
  • parentId: [parent-issue-id]
  • state: "Todo"

Sub-issues to create:

  1. [Sub-task title] — [brief description]
  2. [Sub-task title] — [brief description]

### 0.3 Create git branch ``bash git checkout -b feature/[ISSUE-ID]-[slugified-title] ``

### 0.4 Move issue to "In Progress"

  • Tool: mcp__linear__update_issue
  • id: [issue-id]
  • state: "In Progress"
  • assignee: "me"

## Phase 1: Implementation

### 1.1 [First sub-task title]

  • Files: [files to modify/create]
  • Details: [what to implement]

### 1.2 [Second sub-task title]

  • Files: [files to modify/create]
  • Details: [what to implement]

## Phase 2: Verification

  • [ ] All sub-tasks implemented
  • [ ] Build passes
  • [ ] Manual testing complete
  • [ ] Ready for /create-pr

```

  1. Exit plan mode using ExitPlanMode.
  1. After plan approval: execute Phase 0 in order (0.1 → 0.2 → 0.3 → 0.4), then display the Step 5 summary before starting Phase 1.

Step 4: Update Linear (Simple Features)

For simple features that skipped plan mode:

Use mcp__linear__update_issue:
- id: [issue-id]
- description: [refined description with requirements]
- labels: [type label, priority label, size label]
- priority: [0-4]
- state: "Todo"

Then create branch and move to In Progress:

git checkout -b feature/[ISSUE-ID]-[slugified-title]
Use mcp__linear__update_issue:
- id: [issue-id]
- state: "In Progress"
- assignee: "me"

Step 5: Display Summary

## Feature Ready: [Issue ID] — [Issue Title]

**Branch**: feature/ABC-XX-issue-title (created and checked out)
**Linear Status**: In Progress
**Assignee**: You

**Updated Description**:
[refined description summary]

**Sub-tasks Created** (if applicable):
- ABC-XX: [sub-task title]
- ABC-XX: [sub-task title]

Ready to start implementing! Which sub-task would you like to begin with?

Best Practices

  1. Always refine before starting — Don't jump into code without understanding
  2. Break down large features — Anything > 1 day should have sub-tasks
  3. Add acceptance criteria — Clear definition of "done"
  4. Estimate realistically — Better to overestimate than under
  5. Link related issues — Reference similar work or dependencies

Integration with Other Skills

  • After refinement: Branch is created — just start coding
  • For status checks: /new-feature doubles as a status checker
  • When done: /create-pr to create pull request
  • Check backlog size: /backlog to maintain healthy backlog

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.