IssueLoop closes the gap between a failing test and an actionable bug ticket. It runs your test suite, sends failures to an LLM for triage, deduplicates and prioritizes them, then dispenses one ticket at a time to whoever is fixing bugs — whether that’s a developer, an agent, or an automated pipeline.Documentation Index
Fetch the complete documentation index at: https://mintlify.com/onenot8/issueLoop/llms.txt
Use this file to discover all available pages before exploring further.
Quickstart
Go from install to your first ticket in under five minutes.
CLI Reference
Every command, subcommand, and flag documented with examples.
Python API
Full reference for all 45+ public Python functions.
Live Monitoring
Watch a running process or log file and auto-create tickets from errors.
How IssueLoop works
IssueLoop follows a simple pipeline from raw test output to an actionable, prioritized ticket queue.Scan and run tests
Point IssueLoop at a repo. It reads your
test_manifest.json, runs each test command, and captures stdout/stderr with exit codes.LLM triage
Failing runs are sent to an LLM (Ollama locally, or Anthropic/OpenAI in the cloud). The model splits a single failure run into multiple independent bug tickets, each with a priority label.
Ticket queue
Tickets are stored in SQLite (or Supabase) and dispensed one at a time via
get_top_error(). Claiming a ticket marks it in_progress — no race conditions.Key capabilities
Batch test workflow
Run tests, triage failures, and drain the ticket queue in a single automated pipeline.
Fix-apply layer
Propose fixes and apply them under an explicit, auditable permission allowlist.
LLM providers
Configure Ollama, Anthropic, or OpenAI with automatic priority fallback.
Node.js & Go clients
Use IssueLoop from any language via the local HTTP bridge.
IssueLoop works out of the box with zero configuration. Run
pip install issueloop and the default Ollama backend (local, free) is ready to go.