Observed Signal · Jun 9, 2026 · Technical Release · Source: DEV Community · Impact: 2/5 · Sentiment: Positive
infrawise adds infrastructure visibility for AI coding assistants
infrawise is an open-source developer tool that statically analyzes a project's codebase, DynamoDB tables and PostgreSQL schemas and exposes that infrastructure context to AI coding assistants (demonstrated with Claude Code) via MCP tools. By surfacing table sizes, GSIs, partition keys, existing query patterns and flagged high-severity findings, infrawise enables assistants to make informed, deterministic recommendations (e.g., use Query against a GSI instead of a costly Scan). The project provides a CLI installer, generates an infrawise.yaml and .mcp.json for editor integration, bundles 15 MCP tools for assistants to query, caches analysis for 24 hours, and offers read-only Postgres connection setup instructions and report export flags.
A developer-focused technical release that improves LLM-assisted coding by adding deterministic infrastructure visibility; useful for engineering teams but not industry-shifting for AdTech/MarTech at large.
Track Amazon 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
- infrawise statically analyzes code, DynamoDB tables and PostgreSQL schemas and exposes the context to editors via MCP tools.
- When connected, Claude Code can query 15 MCP tools from infrawise to get concrete findings (e.g., full table scan, missing index, hot partition risk).
- infrawise's analysis is deterministic and model-free, using TypeScript AST parsing (ts-morph), schema introspection, rule-based analyzers and graph correlation.
- Installation and usage examples include: `npm install -g infrawise`, `infrawise start --claude`, an auto-generated `infrawise.yaml` and `.mcp.json`, and CLI analysis/export commands (`infrawise analyze --severity high --output report.md`).
- PostgreSQL integration uses a suggested read-only user with GRANT CONNECT/USAGE/SELECT; analysis cache is refreshed within sessions and persisted for 24 hours.
Connected Companies & Entities
3 Entities mappedOntology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Infrawise: CI Checks to Block AI-Generated Infra Mistakes
The article introduces infrawise check, a CI build-step that analyzes a codebase alongside live infrastructure (DynamoDB schemas, PostgreSQL indexes, Lambda usage) and fails the build when AI-generated or other code would introduce infrastructure anti-patterns such as full table scans or missing indexes. infrawise provides blocking findings with table- and caller-specific details, a severity gating flag (--fail-on high|medium|low) for CI integration (example shown for GitHub Actions), and an infrawise.yaml configuration that supports environment-variable substitution and scoped table analysis. The tool requires read-only AWS permissions (e.g., dynamodb:ListTables, dynamodb:DescribeTable) and is intended to catch regressions that static analysis, tests, and code review cannot detect because they lack live infrastructure context.
New Relic Launches AI Coding Observability
New Relic announced the development of an open-source feature called New Relic AI Coding Observability, designed to extend production-grade monitoring into the AI-assisted coding phase. The capability normalizes telemetry across multiple coding assistants (including Claude Code, Cursor, GitHub Copilot, Windsurf and Amazon Q) and correlates that data with existing production infrastructure. Key functions highlighted include visibility into AI-driven code actions, cost tracking and forecasting, productivity measurement, security and compliance via a local-only/zero-outbound mode, and vendor-neutral interoperability using OpenTelemetry and the Model Context Protocol (MCP). New Relic positions the feature to help engineering and platform leaders govern, audit and optimize AI coding assistant usage. The announcement includes a quote from New Relic Chief Product Officer Brian Emerson.
AI-Assisted SQL Development Using Claude Code
This technical blog explains how Claude Code (Anthropic's AI coding agent) can be used for AI-assisted SQL and ETL development by enforcing project conventions and encapsulating recurring workflows. It describes three practical levers — rules files (.claude/rules/) that make style guides machine-enforced, skills (invoked as slash commands) to reproduce recurring tasks, and agents that orchestrate multi-step ETL workflows. The article argues that machine-enforced conventions plus human review are required to keep generated SQL readable and correct, and it links to example conventions, derivation workflows, and a starter kit repository on GitHub. Publication date: 2026-07-07.
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.