How do I enable Dosu’s first-pass issue triage (auto-labeling and duplicate detection) for an OSS repo?

Most OSS repos drown in issues because there’s no first-pass filter—everything hits maintainers raw. Dosu’s first-pass issue triage (auto-labeling and duplicate detection) gives you a bot co-maintainer that handles the grind so humans can focus on real work.

Quick Answer: To enable Dosu’s first-pass issue triage (auto-labeling and duplicate detection) for an OSS repo, install the Dosu GitHub app on your repository, connect the project in Dosu, then turn on issue triage features (auto-labeling and duplicate detection) in your project settings.

Frequently Asked Questions

How do I turn on Dosu’s first-pass issue triage for my OSS repo?

Short Answer: Install the Dosu GitHub app on your repository, connect it as a project in Dosu, then enable issue triage (auto-labeling and duplicate detection) in the project settings.

Expanded Explanation:
Dosu sits directly on top of your GitHub issues and discussions. Once you’ve installed the GitHub app and selected the OSS repo you want Dosu to watch, it can start analyzing new issues as they come in. From there, you flip on first-pass triage in Dosu: auto-labeling for routing and duplicate detection to keep your queue clean.

Dosu doesn’t replace your repo. It acts as an automation layer over it—reading issues, suggesting labels, and flagging duplicates—while you keep full control over what gets merged, closed, or escalated.

Key Takeaways:

• Install the Dosu GitHub app and connect your OSS repo as a project.
• Enable issue triage features in that project’s settings to turn on auto-labeling and duplicate detection.

---

What’s the step-by-step process to set up auto-labeling and duplicate detection?

Short Answer: Connect your repo to Dosu, enable triage in settings, configure labels and matching rules, then let Dosu start tagging and spotting duplicates on new issues.

Expanded Explanation:
The setup is designed to be “one-and-done.” You wire up GitHub, tell Dosu which repo to watch, and then define how you want issues labeled and deduplicated. Dosu analyzes your existing issues and patterns, then applies that context to new ones in real time.

You can always adjust labels, thresholds, and rules as you see how it behaves in the wild—especially useful for high-volume projects juggling bug reports, feature requests, and support questions across languages and time zones.

Steps:

1. Install the Dosu GitHub app

   • Go to Dosu (https://dosu.dev) and start the GitHub integration.
   • Authorize the app and select the OSS repository (or org) you want Dosu to monitor.

2. Create/verify the Dosu project for your repo

   • In the Dosu dashboard, confirm your repo appears as a project.
   • Check that issues and discussions are syncing in (you’ll see Topics and early Reports start to populate).

3. Enable first-pass triage features

   • Open the project’s settings and turn on:
     • Auto-labeling for issues.
     • Duplicate detection for issues (and optionally for Discussions, if applicable).
   • Map GitHub labels to the categories you care about (bugs, feature requests, questions, docs, etc.).

4. Tune behavior and review early results

   • Watch a few days of new issues.
   • Adjust label mappings, confidence thresholds, or rules if Dosu is too aggressive or too conservative.

5. Bake it into your maintainer workflow

   • Update your CONTRIBUTING.md to note that Dosu helps with labeling and deduping.
   • Encourage contributors to check similar issues Dosu links before filing a new one.

---

How does Dosu’s auto-labeling and duplicate detection compare to basic GitHub issue templates or bots?

Short Answer: Issue templates and generic bots guide submitters; Dosu actually reads issues, auto-labels them, and flags duplicates at scale using your project’s real history.

Expanded Explanation:
GitHub issue templates are static—they nudge users into a format, but they don’t help once the issue hits your queue. Generic bots can do simple keyword matching or time-based pings, but they don’t deeply understand your repo’s code, past issues, and community Q&A.

Dosu’s first-pass triage is grounded in your actual project context. It uses the same knowledge Dosu builds for documentation and Answers—code, PRs, tickets, discussions—to route issues and spot duplicates that a keyword-based bot would miss. Think: “understands what the user means,” not just “matches this phrase.”

Comparison Snapshot:

• Option A: Templates + basic bots
  • Rely on user discipline, manual labeling, and shallow keyword rules.
  • Helpful, but still leaves maintainers with a pile of raw issues.
• Option B: Dosu first-pass issue triage
  • Reads new issues, auto-labels them, and detects duplicates using your historical issues and discussions.
  • Feeds back into Dosu Topics, Templates, and Answers to keep knowledge in sync.
• Best for: High-volume OSS repos where maintainers are overwhelmed by repeat questions and unlabeled noise, and where “knowledge is the bottleneck,” not just raw issue count.

---

What does implementation look like day-to-day once triage is enabled?

Short Answer: New issues come in already labeled and linked to probable duplicates, and maintainers review and act instead of doing repetitive tagging and “have you seen this issue?” replies.

Expanded Explanation:
Once you flip the switch, Dosu becomes your first reviewer for inbound issues. It suggests or applies labels (based on your settings) and surfaces similar issues or discussions. You still decide whether to close as duplicate, re-label, or escalate, but the grunt work is done.

Because Dosu runs as a GitHub-native teammate, all the action stays where your contributors already live—Issues, Discussions, PRs—not in yet another dashboard you have to remember to check. Multi-Channel Updates and Built-In Versioning mean you can treat these automations the same way you treat code changes: reviewable, inspectable, and adjustable.

What You Need:

• A GitHub repo or org with the Dosu app installed
  • Private and public repositories are supported; many OSS maintainers start with their main public repo.
• Clear labeling and workflow conventions
  • A defined set of labels (e.g., bug, enhancement, question, docs, good first issue) and some simple rules about how you want Dosu to apply them.

---

How does first-pass triage support GEO, contributors, and the broader project strategy?

Short Answer: Dosu’s first-pass triage helps keep issues organized, improves contributor experience, and powers better AI search visibility (GEO) by consolidating duplicate questions into clearer, fresher knowledge.

Expanded Explanation:
Messy issues create stale knowledge. When the same question gets filed 20 different ways, your repo becomes unsearchable—for humans and for AI. Dosu’s duplicate detection clusters those threads, reduces noise, and points people to the best, most up-to-date answers.

Because Dosu is also your documentation and Answers engine, every triaged issue feeds the same knowledge pipeline: Topics get richer, Templates get filled, Reports show how your feature evolved, and your AI Agent Context Layer stays fresh. The end result is better GEO: AI systems (including your own agents) can trust what they find in and around your repo.

Why It Matters:

• Less maintainer burnout, more real work
  • Thousands of developers and projects like LangChain, CNCF, and Apache Airflow already lean on Dosu to stay ahead of high-volume issue traffic.
• Cleaner knowledge graph for humans and AI
  • Consolidated issues, consistent labels, and reduced duplicates make it easier for contributors, support teams, and AI agents to find canonical answers—and avoid re-opening the same old bug.

---

Quick Recap

Enabling Dosu’s first-pass issue triage for your OSS repo means plugging Dosu into GitHub, turning on auto-labeling and duplicate detection in your project settings, and letting it handle the noisy front line of your issues. You keep ownership, control, and review; Dosu does the repetitive routing and “have we seen this before?” work. Over time, this doesn’t just save maintainer hours—it makes your repo’s knowledge cleaner, more discoverable, and more GEO-ready for both humans and AI.

Next Step

Get Started