SShortSingh.
Back to feed

Developer finds 93% of fully-signed Safe multisig transactions are permanently unexecutable

0
·4 views

A developer building an automated executor for Safe multisig transactions ran a read-only detector against 1,299 Base mainnet Safes before writing any code, scanning blocks from Base's early history through block 50,776,046. The scan identified 366 fully-signed but unexecuted transactions, yet 339 of them turned out to be permanently dead — roughly 93% of the total. The root cause is that Safe executes nonces strictly in order, so if a different transaction consumes a given nonce, the original remains in the queue indefinitely with no indication it can never run. The naive one-line detector missed this entirely because it never compared queued nonces against the Safe's live on-chain nonce, a gap only exposed by an additional RPC call. This finding shifted the project's focus: instead of an execution engine, the core product became a structured refusal system that explains precisely why a transaction cannot or should not be submitted.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

Developer Builds Self-Hosted Ticket System to Test Real-World AI Coding Agents

A developer created Lutions, a self-hosted web application for project and ticket management, partly driven by rising software licence costs and concerns over digital sovereignty. The project grew beyond its original scope into a personal laboratory for testing agentic AI in a realistic, mature codebase rather than a simplified demo environment. The developer argues that useful AI coding agents require more than better prompts — they need access to a project's full context, conventions, history, and constraints. Drawing parallels to benchmarks like SWE-bench, the article highlights that real-world agentic development demands agents navigate multi-file changes, existing rules, and organisational expectations. The piece frames Lutions as a field study in what conditions must exist for AI agents to contribute meaningfully to established software projects.

0
ProgrammingDEV Community ·

How Engineering Discipline Can Help AI Generate Production-Quality Code

A software consultant and AI advisor describes a methodology for producing production-ready code using AI without directly writing or reviewing the code during development. The approach stems from a personal experiment he began roughly two years ago, driven by concerns that rapid AI advancement would erode his technical expertise. He argues that while the method consumes more tokens upfront, it reduces rework and lowers the total cost of ownership over time. The technique is rooted in engineering discipline, requiring attention to correctness, security, performance, and observability rather than relying on 'vibe-coding.' He has published a free tutorial and an open-source skills framework on GitHub to help others adopt the approach.

0
ProgrammingDEV Community ·

EU Cyber Resilience Act Mandatory Vulnerability Reporting Begins September 11

The European Union's Cyber Resilience Act (CRA) activates its mandatory incident reporting requirements on September 11, 2026, affecting software makers and device manufacturers selling products in the EU. Under the new rules, companies must issue an early warning within 24 hours of discovering an actively exploited vulnerability, followed by a formal notification within 72 hours and a final report within 14 days. Reports must be submitted to the relevant national CSIRT and to ENISA through a Single Reporting Platform, with the clock starting the moment a team becomes aware of an issue. The regulation targets any manufacturer placing a product with digital elements on the EU market, including apps, firmware, and connected devices, though pure SaaS services generally fall under the separate NIS2 directive. Full CRA compliance, including CE marking, is not required until December 2027, but the reporting obligations take effect immediately this Friday.

0
ProgrammingDEV Community ·

Developer Guide: Building an AI Agent That Completes and Submits Gig Platform Jobs

A developer tutorial published on DEV Community outlines how to build an autonomous AI agent capable of completing paid tasks on freelance platforms like Upwork and Fiverr. The system connects a large language model (LLM) chain — built with tools like LangChain and OpenAI's API — to a gig platform's REST or webhook interface to receive, process, and submit jobs automatically. The guide recommends keeping the agent's scope narrow, such as writing SEO blog posts or converting design files to code, to reduce errors and simplify pricing. Sample Python code is provided for the LLM prompt chain and a FastAPI-based webhook handler that listens for new job postings and returns completed work. The tutorial also addresses cost trade-offs, noting that managed APIs like GPT-4o-mini cost roughly $0.002–$0.01 per call, while self-hosted models offer savings at the expense of greater infrastructure complexity.