backlog-this Command
By Jay Griffin, GPT-5.3-Codex · February 16, 2026
docs
🏷️ Tags:workflowaibacklogproductivitymetaspec
A focused workflow for capturing ideas into backlog storage when I intentionally want to stop implementing and preserve context for later.
Backlog-This Trigger Word Workflow
Why this exists
I can move fast with AI, but speed can turn into scatterbrain mode if I follow every interesting branch immediately. I want a clean interrupt mechanism that lets me preserve good ideas without derailing what I am actively shipping.
backlog-this is that mechanism.
Core concept
When I say backlog-this, I am switching from build mode to capture mode.
- Build mode: implement, edit code, run commands, ship changes
- Capture mode: do not implement; only structure and store the idea for later
Goals
- Avoid losing promising ideas
- Reduce context-switch thrash
- Keep implementation focus on current priority
- Preserve enough detail to resume quickly later
Non-goals
- No immediate feature implementation
- No architecture rabbit holes during capture
- No bikeshedding format details before capture exists
Command
If I say backlog-this, the workflow should:
- Stop proposing implementation steps for the current tangent
- Summarize the idea and current context
- Store it as a backlog entry using a consistent template
- Return to the original active track
Backlog entry template
Each parked idea should capture:
- Title: concise and searchable
- Problem: what pain or opportunity I noticed
- Proposed direction: rough path, not full solution
- Why not now: what current priority this interrupts
- Next step: smallest concrete action for future me
- Context snapshot: relevant project state, files, links, or assumptions
- Tags: classification for later filtering
Storage options
Option A: Single running backlog doc
- Fastest to start
- One place to scan
- Can get noisy over time
Option B: One file per parked idea
- Cleaner long-term organization
- Better linking and metadata
- Slightly more overhead per capture
Recommended starting point
Start with Option A for speed, then migrate to Option B if volume grows.
Success criteria
This workflow is successful if:
- Capturing an idea takes less than 60 seconds
- I can reliably find parked ideas later
- Parked ideas are actionable when reopened
- I stay focused on the current shipping task
Next refinement pass
In the next pass, I want to lock:
- Exact storage shape
- Exact field schema
- Retrieval commands (e.g., list open, list recent, reopen by slug)