AI AGENTS

Enrich criteria-less stale issues before sprint planning

Weekly, an agent finds backlog issues older than 30 days that lack acceptance criteria, drafts criteria from the title and description.

CategoryAI Agents
Enginepaperclip
Difficultybeginner
Triggerschedule
Steps6
Setup~5 min

How it runs

The automated pipeline, trigger to output.

  • TriggerWeekly schedule starts the run
  • ActionFind 30+ day issues with no criteriaLinearLinear
  • ActionDraft acceptance criteria per issueOpenAI
  • LogicRoute confident drafts vs needs-info
  • ActionComment drafted criteria on each issueLinearLinear
  • OutputSend Slack digest of enriched + needs-infoSlack

What it does

This agent attacks backlog rot. It finds issues that have sat untouched for over a month with no acceptance criteria — the items that stall every refinement session — and drafts concrete, testable criteria from whatever context the title and description provide. It never overwrites; it posts the draft as a comment so an owner can accept or edit.

When to use it

Use it on long-lived backlogs where unrefined issues pile up and refinement meetings burn time writing criteria from scratch. Running it weekly means the team always has a stack of pre-drafted criteria ready to review.

How it works

  1. 1A weekly schedule starts the run.
  2. 2The agent queries Linear for issues older than 30 days with empty acceptance criteria.
  3. 3An LLM drafts acceptance criteria for each, marking any issue where context is too thin to draft confidently.
  4. 4A logic step routes confident drafts to enrichment and low-context issues to a needs-info list.
  5. 5The agent posts drafted criteria as a comment on each confident issue.
  6. 6A Slack digest lists enriched issues plus the needs-info items that require a human to add context.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect LinearIssues, projects, cycles, triage.
  2. 2
    Connect OpenAIModels, embeddings, files.
  3. 3
    Connect SlackChannels, DMs, threads, mentions.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.