cd /news/ai-startups/novee-launches-agentic-fix-into-codi… · home topics ai-startups article
[ARTICLE · art-14467] src=letsdatascience.com pub= topic=ai-startups verified=true sentiment=↑ positive

Novee launches Agentic Fix into coding assistants

AI penetration-testing startup Novee launched Agentic Fix, a capability that routes validated exploit findings and remediation guidance directly into developer coding assistants such as Claude, Codex, Copilot, Cursor and Devin. The tool generates a GitHub issue from exploit context, has the selected agent produce a fix and open a pull request, and reassesses the asset after the fix is merged to confirm the vulnerability is resolved. Novee, which launched with $51.5 million in funding, aims to eliminate bottlenecks between security and engineering teams.

read3 min publishedMay 26, 2026

According to SiliconANGLE, AI penetration-testing startup Novee launched Agentic Fix, a capability that pushes validated exploit findings and remediation guidance directly into developer coding assistants. SiliconANGLE reports Agentic Fix generates a detailed GitHub issue from the validated exploit context, routes that guidance to chosen coding agents including Claude, Codex, Copilot, Cursor and Devin, and has the selected agent produce a fix and open a pull request. After a fix is merged, SiliconANGLE reports Novee reassesses the affected asset to confirm the original vulnerability is resolved. SiliconANGLE quotes co-founder and CEO Ido Geffen saying, "We're bringing security and engineering teams into the same loop and eliminating bottlenecks." SiliconANGLE also reports Novee launched with $51.5 million in funding.

What happened

According to SiliconANGLE, AI penetration-testing startup Novee launched a new capability called Agentic Fix that routes validated exploit findings and remediation guidance into developer coding assistants. SiliconANGLE reports the capability uses the exploit context that uncovered a vulnerability to generate a detailed GitHub issue and remediation guidance. Per SiliconANGLE, Novee then routes that guidance to the customer-selected coding agents, which the article lists as Claude (Anthropic), Codex (OpenAI Group), GitHubs Copilot, Cursor and Cognition AIs Devin.

Technical details

According to SiliconANGLE, the workflow reported by Novee creates a fix by having the chosen coding agent produce code and open a pull request against the repository, and Novee reassesses the affected asset after a merge to confirm the original vulnerability is resolved. SiliconANGLE quotes Ido Geffen, co-founder and CEO, saying, "We're bringing security and engineering teams into the same loop and eliminating bottlenecks."

Editorial analysis - technical context

Companies and teams that integrate security tooling with developer workflows often aim to reduce handoff latency between discovery and remediation. Industry-pattern observations note that moving validated exploit context into tools already embedded in engineering workflows can reduce time spent on triage and prevent superficial fixes that do not address root causes.

Context and significance

Reporting frames Agentic Fix as targeting a rising bottleneck where autonomous testing shortens discovery timelines but leaves triage, assignment, patching and retesting largely manual. For practitioners, automated handoff into code assistants combined with post-merge reassessment can change how security findings are prioritized and validated in CI/CD pipelines, increasing the importance of test gating and reproducible exploit context.

What to watch

Observers should track adoption indicators such as which coding assistants customers prefer, integration patterns with existing CI/CD and issue-tracking systems, and whether automated PRs increase the rate of validated fixes versus reverted or ineffective patches. SiliconANGLE reports Agentic Fix is available immediately, and notes Novee previously launched with $51.5 million in funding.

Note: All product details, integrations, quotes and funding figures above are reported by SiliconANGLE. Novee has not been quoted in additional sources in the materials reviewed here.

Scoring Rationale #

This is a notable product release for security and engineering teams: integrating validated pentest output with developer coding assistants can materially shorten remediation cycles. The story is product-level rather than a frontier research breakthrough, so it rates as a mid-tier but relevant development for practitioners.

Practice interview problems based on real data

1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.

Try 250 free problems

── more in #ai-startups 4 stories · sorted by recency
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/novee-launches-agent…] indexed:0 read:3min 2026-05-26 ·