Jira Automation vs Workflow Validators
If a check in Jira Cloud has to stop a work item from moving, put it on the workflow transition as a...
Tag archive
If a check in Jira Cloud has to stop a work item from moving, put it on the workflow transition as a...

Remove inactive users in Jira Cloud by a rule in plain words: a CogniRunner agent ranks them by last active date and a person approves each removal.

LLM regression testing for a Jira workflow validator: 8 labelled issues, a Node script and CogniRunner's Rules API, run before you change the model or prompt.

Read the Jira API rate limit headers on your own call with curl, then see which CogniRunner component spends the hourly quota and when Atlassian throttled it.

A Jira coding agent that reads your GitHub or Bitbucket repo from the issue, stages edits, and commits to a branch or opens a PR once you confirm the file list.

LeanZero's first year: a portfolio headlined Things I've Built became four Forge apps on the Atlassian Marketplace, and CogniRunner had its busiest two months.

Connect Jira to OpenAI, Claude, Gemini, Bedrock or the zero-key Forge LLMs behind one CogniRunner workflow validator, and what bringing your own key changes.

Claude Code moves a Jira change request through its workflow via the REST API. Two CogniRunner validators read the backout plan and rehearsal: 400, then 204.

Jira LLM integration with a local model: pin one Jira project to LM Studio over a Tailscale Funnel with CogniRunner, while the site keeps its cloud AI provider.

CogniRunner now runs inference on LM Studio locally or zero-key on the Forge LLM provider, plus plain-English AI actions that write back to Jira.

Call Atlassian-hosted Forge LLMs or your own provider via declared egress, then wire the model into Jira validators, conditions and post-functions.

A condition takes only an expression, so its function key is ignored and your handler never runs while forge lint and deploy pass. Two fixes measured.