frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: IncidentFox, AI SRE that auto-builds its own integrations (open source)

https://github.com/incidentfox/incidentfox
1•chiehminwei•1h ago
We've been building an AI agent that debugs production incidents, and we think most AI SRE tools are solving the wrong problem.

Everyone's focused on the reasoning — better prompts, better models, RAG over runbooks. But in our experience, the bottleneck isn't the AI's ability to think. It's access to data. At any company past 50 engineers, half the context you need during an incident lives in internal tools with no public docs, custom deploy systems, homegrown dashboards, internal CLIs. No vendor integration covers this.

So we focused on the integration problem. On setup, IncidentFox reads your codebase and Slack history and auto-generates tool integrations — including for internal tools it's never seen before. It figures out what APIs exist, how they're called, and builds the connectors. After each incident, it learns what data sources were actually useful and refines.

The agent itself is straightforward: alert fires → form hypotheses → query observability stack → correlate → report findings in Slack. For example, if latency spikes on your payments service, it'll pull the deploy from 20 minutes ago, find the relevant code change, check downstream dependencies, and open a PR with a fix.

Three design decisions we're opinionated about:

1. Slack-native, not dashboard-native. At 3am you don't want another tab. Paste screenshots, drop log files, view traces — all in the thread.

2. Team-customizable agents. The SRE team's agent should behave differently from the platform team's. Engineers can configure prompts and build team-specific tools. One-size-fits-all doesn't work for incident response.

3. No manual MCP server setup. If you need to spend weeks configuring integrations before AI is useful, you've already lost.

Apache 2.0, fully self-hostable. Try it in our Slack (https://join.slack.com/t/incidentfox/shared_invite/zt-3ojlxv...) with real telemetry (no setup), or clone and run it yourself.

The technical piece we find most interesting: given enough incidents, the agent builds an emergent model of your entire infrastructure — which services are fragile, which deploys are risky, which alerts are noise. Nobody explicitly teaches it this. It just falls out of the investigation data. We think there's something deeper here about AI agents that learn system topology from failure patterns.

Would love to hear people's thoughts!

IPlotCSV: CSV Data, Visualized Beautifully

https://www.iplotcsv.com/demo
1•maxmoq•14s ago•1 comments

Go's Synctest Is Amazing

https://oblique.security/blog/go-synctest/
1•thunderbong•35s ago•0 comments

Firesheep: Easy HTTP session hijacking from within Firefox (2010)

https://codebutler.com/2010/10/24/firesheep/
1•lelandbatey•50s ago•0 comments

The Data Pipeline for Superintelligence Starts with Your Screen

1•Nadav--Shanun•2m ago•0 comments

Are OpenClaw and AgentSkills Safe?

https://openclaw.ai/
1•doanbactam•3m ago•1 comments

Slag Field Ecology

https://johncarlosbaez.wordpress.com/2026/02/03/slag-field-ecology/
3•OgsyedIE•3m ago•0 comments

Malicious Skills Found in OpenClaw's ClawHub Marketplace

https://www.esecurityplanet.com/threats/hundreds-of-malicious-skills-found-in-openclaws-clawhub/
2•insiderllm•4m ago•1 comments

Ask HN: Wen Moltbook for Customer Service?

2•verdverm•4m ago•0 comments

How to Program Computers (KOS)

https://fall.tr/kos
2•todsacerdoti•6m ago•0 comments

Number Trees: Interactive explorer of the integer-tree bijection

https://echomoltinsson.github.io/number-trees/
1•onyx_writes•8m ago•0 comments

Claude has been having a moment – can it keep it up?

https://www.theverge.com/report/874308/anthropic-claude-code-opus-hype-moment
2•ecliptik•9m ago•0 comments

$50 on war in Iran: How geopolitical bets have surged on Polymarket

https://restofworld.org/2026/polymarket-online-betting-politics-war-charts/
1•i7l•10m ago•0 comments

Intentionally Designing the Future of AI

https://www.goodfire.ai/blog/intentional-design
1•jal278•10m ago•0 comments

Show HN: Lytok – A binary format with 44% fewer tokens than JSON

https://github.com/Joguel96/lytok-spec
2•joguel96•14m ago•0 comments

Flow-Based Programming: Seminal Texts and Theoretical Foundations

https://repolex.ai/blog/2025/12/10/fbp-seminal-texts/
2•arto•18m ago•0 comments

BMW Commits to Subscriptions Even After Heated Seat Debacle

https://www.thedrive.com/news/bmw-commits-to-subscriptions-even-after-heated-seat-debacle
3•geox•24m ago•0 comments

Show HN: Similar Repos – AI-Powered Recommender for GitHub Repository

https://github.com/SimilarRepos/similar-repos
2•robbyzhao•32m ago•0 comments

1M token context: The good, the bad and the ugly (2025)

https://www.micron.com/about/blog/company/insights/1-million-token-context-the-good-the-bad-and-t...
1•teleforce•33m ago•0 comments

Making AI models leaner and faster without sacrificing accuracy

https://research.google/blog/sequential-attention-making-ai-models-leaner-and-faster-without-sacr...
1•Anon84•34m ago•0 comments

The Software Rout Is Spreading Pain to the Debt Markets

https://www.wsj.com/finance/investing/the-software-rout-is-spreading-pain-to-the-debt-markets-d6d...
4•petethomas•36m ago•0 comments

Entwinings: Literature and History, Fathers and Sons, Writers and Readers

https://hedgehogreview.com/issues/place-and-revolution/articles/entwinings
1•Caiero•37m ago•0 comments

CISA gives federal agencies one year to rip out end-of-life devices

https://therecord.media/cisa-gives-federal-agencies-one-year-end-of-life-devices
2•WaitWaitWha•39m ago•0 comments

Bitcoin gets a zero price target in wake of Burry warning

https://seekingalpha.com/news/4547997-bitcoin-s-price-target-at-zero
8•hn_acker•39m ago•0 comments

YARN: Efficient Context Window Extension of Large Language Models (2024) [PDF]

https://proceedings.iclr.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Co...
1•teleforce•40m ago•0 comments

Bios over SSH: Turning KVM Video into Live Text (No Serial Needed)

https://www.youtube.com/watch?v=lhBzCPXqWAA
2•thunderbong•40m ago•0 comments

'Depths of Wikipedia' Creator Annie Rauwerda on 'Fragile' Internet Citations

https://blog.archive.org/2026/02/05/depths-of-wikipedia-creator-annie-rauwerda-on-fragile-interne...
1•abetusk•42m ago•0 comments

Integration Labs: Transforming Traumatizing into Trauma-Integrating Cultures

https://pocketproject.org/integration-labs-2026/
1•rendx•43m ago•0 comments

I Hacked My Feed to Find Undervalued Content

https://xrayfeed.deepwalker.xyz/blog/strategies
1•m00dy•44m ago•0 comments

What are tokens and how to count them?

https://help.openai.com/en/articles/4936856-what-are-tokens-and-how-to-count-them
1•teleforce•47m ago•0 comments

Web of Knowledge

https://dave.autonoma.ca/blog/2019/06/06/web-of-knowledge/
1•andsoitis•49m ago•0 comments