Observed Signal · Jul 13, 2026 · Technical Release · Source: DEV Community · Impact: 1/5 · Sentiment: Positive
AI Converts GitHub Issues Into Tested Pull Requests
resolvo is an agentic pipeline built by Krishna Khandelwal that converts a GitHub issue plus a repository URL into a working, tested pull request without requiring a local clone. Implemented with LangGraph, the pipeline routes work by confidence (fast-track, standard, critical), executes tests in an E2B sandbox using pytest-json-report, and uses a multi-signal retrieval stack (BM25, Cohere rerank, symbol matching) fused via Reciprocal Rank Fusion. High-stakes planning and adversarial review use Gemini Flash models while lighter per-file work uses Google's lite models; Gemini calls are grounded with live Google Search results. A demo video and the project repository are linked in the article. The piece was published on 2026-07-13.
Personal technical release of a developer automation tool using LLMs; relevant to developer workflows and LLM applications but not industry-shifting for the AdTech/MarTech ecosystem.
Track GitHub Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- resolvo converts a GitHub issue and a repo URL into a working pull request with tests passing and an automated review.
- The project is built with LangGraph and structured as a StateGraph multi-agent pipeline.
- The pipeline uses Gemini Flash models for high-stakes planning and adversarial code review, and Google's lighter models for per-file implementation and test generation.
- Grounding with Google Search is integrated into Gemini calls to pull live web results for up-to-date API and advisory information.
- Tests run inside an E2B sandbox against a shallow clone and parse results with pytest-json-report; retrieval combines BM25, Cohere rerank-v4.0, symbol matching, and Reciprocal Rank Fusion.
Connected Companies & Entities
4 Entities mapped“resolvo is an agentic pipeline that takes a GitHub issue and a repo URL and hands you back a working pull request — with tests already passi...”
“I used Gemini Flash models for the two critical steps that need the most contextual judgment ... while Google's lite models handle enrichmen...”
“The planner fuses five signals — raw-issue BM25, enriched-query BM25, Cohere rerank-v4.0, symbol-name matching, and one-hop dependency expan...”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Building an Agentic PR Reviewer with Antigravity SDK
Google announced on 2026-06-18 that it is unifying its AI terminal tools by transitioning the community-focused Gemini CLI into Antigravity CLI. This Dev.to post demonstrates how to migrate by building an automated first-pass pull request reviewer using the Google Antigravity SDK and the run-agy-sdk composite GitHub Action. The article describes an agentic review pipeline that runs a managed Antigravity Agent in isolated sandboxes, uses the GitHub MCP server to post PR comments, requires an ANTIGRAVITY_API_KEY secret, and recommends running the SDK on the GitHub Actions host (not inside a container) to allow controlled access to Docker-based MCP servers. The author publishes workflow YAML, security recommendations (restrict triggers to opened/reopened and limit fork execution), and shares the run-agy-sdk repository as a template for teams to adapt.
Build an Autonomous AI Agent to Open GitHub PRs Overnight
A technical how-to describing an architecture for autonomous AI coding agents that convert tasks (e.g., GitHub issues) into reviewable pull requests without human intervention. The author breaks the workflow into five stages — Ingest, Plan, Execute, Verify, Package — and emphasizes chaining narrow, inspectable steps rather than a single large prompt. The guide details GitHub integration best practices (one branch per task, draft PRs, provenance labels, CI checks), security controls (fine-grained personal access tokens, run in disposable containers), operational limits (retry ceilings, token/dollar ceilings), and the kinds of tasks agents handle reliably (mechanical, objectively verifiable changes) versus those they fail at (ambiguous product work or repos with weak test suites). The article reports the pattern was implemented and run against real repositories and offers pragmatic safety and cost recommendations.
Pull Requests Decline as AI Agents and Models Surge
Google has released a native Gemini app for macOS, available worldwide as a free download for devices running macOS 15 or later. The Mac app provides a system shortcut (Option + Space) for instant access and can answer questions using user files and code, supports single-tab screen sharing, and enables on-desktop creative workflows including image generation via Nano Banana 2 and video creation using Google’s Veo 3.1. Separately, Google announced Gemini 3.1 Flash TTS — a text-to-speech model with studio-style audio-tags supporting 70+ languages — and broader availability of Personal Intelligence in Arabic countries for paid tiers (AI Ultra, Pro, Plus), with extended rollouts for Gemini in Chrome planned. Many capabilities are available via Google Vids and as previews in the Gemini API. (Article published 2026-04-16 by Niklas Lewanczik / OnlineMarketing.de, reproduced on t3n.)
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.