Launching today
Osloq
An AI agent that reproduces GitHub issues for you
177 followers
An AI agent that reproduces GitHub issues for you
177 followers
Most AI dev tools just read your code and guess. Osloq actually runs it. Connect your GitHub, pick an issue, and an AI agent spins up a real sandbox, clones your repo, runs it, and tries to reproduce the bug the way a developer would. You get a report backed by real evidence. What happened, the steps it took, and whether the bug is real, not a hallucinated guess. No local setup, no "works on my machine." It handles the tedious reproduction step so you jump straight to fixing.
How does it handle repos that need specific env vars or a database to actually run, does it ask me to configure those or just spin up a blank sandbox and fail?
@mustafarz0m hey! good question, this is the part we put the most work into. you can add per-repo secrets and env vars in the dashboard (encrypted, only decrypted inside the sandbox) so anything the repo needs like api keys or DB urls is loaded into the environment when the run starts, and when something isn't configured it doesn't just fail on a blank box, the agent spins up a local stand-in (like a local postgres) so it can still run the code path and reproduce the bug instead of giving up.
Really interesting approach!!!!
Curious....how does Osloq handle issues that depend on external services or environment specific configurations?
Anyways...congrats on the launch! 🚀
@worksforme hey, thanks! you can add per-repo env vars and secrets (encrypted), and when an external service isn't reachable it falls back to a local stand-in (like a local postgres) so it still hits the code path instead of dying on a missing dependency.
Reproducing the issue is genuinely the worst half of the job - half the tickets I get are 'it doesn't work' with zero context, and you burn an hour just recreating the state before you can even start fixing. Automating the boring forensic part instead of the fix itself is a smart wedge. How does it handle issues that need specific data/state to trigger? Congrats on the launch.
@david_marko hey, thanks! and yeah, that "it doesn't work" with zero context is exactly the pain. for data/state, the agent sets it up itself, it writes and runs scripts to seed the db or force the exact state it needs, then reproduces from there. when the issue has no repro steps it also digs through the code to infer what state would actually trigger the bug.
Neat concept. Does it handle bugs that need specific user state or production data to actually reproduce?
@dhiraj_patel5 hey, thanks! it never touches your production data, the agent recreates the needed state inside the sandbox itself. you can also add project secrets (encrypted) if the repo needs real config. and if the state truly can't be recreated, it says so instead of guessing.
Curious how it handles repos with external service dependencies like a database or third party API. Does the sandbox spin those up too, or do I need to provide credentials before it can run anything?
@evren8h3u hey, no credentials needed to start. the agent figures out what the repo depends on and stands in for it inside the sandbox, a local database instead of your hosted one, a mock instead of a third party API. if the bug genuinely needs the real service you can add encrypted project secrets and it uses those. either way it runs, missing credentials don't block the investigation
The evidence-backed part is the important line here. For a small team, a bug-repro agent is useful only if the report can become a reviewable artifact: exact assumptions, commands, env gaps, and ideally a tiny failing test or script someone can rerun. That keeps it from turning into another confident AI opinion in the triage queue.
the smartest thing here isn't the sandbox, it's what you DIDN'T build. every other AI dev tool promises to fix the bug, you scoped to proving it exists, which is the one step where "confident hallucination" gets caught instead of shipped
"if it can't back a claim with evidence, it says so" should honestly be the bar for the whole agent category
@yarslav this might be my favorite comment of the launch. yeah, the restraint was the whole point, proving a bug exists is where hallucination gets caught before it ships. and completely agree, "no evidence, no claim" should be the bar for the category