FAQ

Frequently asked questions about Slopcannon.

General

What is a "slop score"?

The slop score is a numeric value representing the amount of potentially problematic patterns detected in a pull request. Higher scores indicate more issues found. The score is additive—each finding contributes points based on its severity.

Is Slopcannon anti-AI?

No! Slopcannon isn't against AI-assisted coding. It's a tool to help teams maintain quality when using AI tools. AI assistants are incredibly useful, but their output sometimes needs human refinement. Slopcannon helps identify those cases.

Does Slopcannon post comments on my PRs?

By default, no. Slopcannon runs in "ghost mode" which means it analyzes PRs and stores results in the dashboard, but doesn't post any comments or create check runs on GitHub. This lets you evaluate the tool without impacting your workflow.

What languages are supported?

Slopcannon analyzes diffs, so it works with any programming language. However, the heuristic detection is currently optimized for:

• JavaScript / TypeScript
• Python
• Go
• Ruby
• Java
• C# / .NET
• PHP

Other languages will still be analyzed but may have fewer specific pattern detections.

Scoring

Why is there no maximum score?

The slop score is additive by design. A PR that touches many files with many issues should score higher than one with a single problem. This provides a more accurate representation of the review effort needed.

What's a "good" score?

However, context matters. A large refactor might naturally have a higher score than a small bug fix.

How is the threshold used?

The threshold (default: 30) determines which PRs are categorized as "flagged" vs "clean" in statistics. It doesn't block or reject PRs—it's purely for reporting.

Privacy & Security

What data does Slopcannon access?

Slopcannon accesses:

• Repository metadata (names, visibility)
• Pull request details (title, author, state)
• Pull request diffs (code changes)
• Organization member list (for team features)

Is my code stored?

Slopcannon stores:

• PR metadata (title, number, author)
• Analysis findings (file path, line number, code snippets for context)
• Scores and statistics

Full repository contents are NOT stored. Diffs are fetched, analyzed, and then only findings are persisted.

Can I delete my data?

Yes. Uninstalling the GitHub App will mark your installation as removed. Reach out to [email protected] for complete data deletion.

Troubleshooting

PRs aren't being analyzed

Check the following:

1. Repository enabled? Check Dashboard → Repos
2. Draft PR? Drafts may be skipped based on settings
3. Recently installed? Initial sync may take a few minutes
4. Webhook working? Check GitHub App settings → Advanced → Recent Deliveries

Scores seem inconsistent

Slopcannon analyzes the diff, not the full file. A PR that touches problematic existing code won't be flagged unless the PR itself changes those lines.

Historical PRs are missing

Historical import covers the last 90 days of closed PRs. PRs older than that aren't imported. Open PRs are always imported regardless of age.

Can I re-analyze a PR?

Currently, re-analysis requires pushing a new commit to the PR. Automatic re-scan functionality is planned for a future release.

Billing & Limits

Is Slopcannon free?

Slopcannon is currently in beta and free to use. Future pricing will be announced before the beta period ends.

Are there rate limits?

Slopcannon respects GitHub API rate limits. For very large organizations with many concurrent PRs, analysis may be slightly delayed during peak times.

How many repositories can I analyze?

There's no hard limit on repositories during beta. Install on as many repos as you'd like.

Feature Requests

Will you add X detection pattern?

Probably! The detection heuristics are actively being improved. Reach out to [email protected] with specific patterns you'd like detected.

Can I customize the scoring weights?

Not currently. Custom scoring rules are on the roadmap.

Will there be IDE integration?

We're considering VS Code and JetBrains integrations. Reach out to [email protected] if this would be valuable for your team.