AgentProto 0.5.0 Adds Credential Brokering, Sandboxes, and Honest Cost Tracking

·5 views

AgentProto released version 0.5.0 on July 8, 2026, shipping 37 packages across six new modules installable via npm. The update introduces a credential broker that silently refreshes auth headers before expiry and a companion secrets package that forwards sensitive tokens to child processes without exposing them in environment files. A new sandbox module defines a provider interface with an E2B-backed implementation, allowing full agent sessions to run in isolated environments with storage sync via GitHub. Cost accounting now tags every usage event with a four-way source label — distinguishing real reported costs, computed prices, missing catalog entries, and unmeasured sessions — so no cost figures are fabricated. Additional updates include an evaluation runner with pluggable LLM judges, an OpenTelemetry telemetry adapter, and a zero-dependency redaction utility that catches secret patterns by value shape rather than key name.

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)

How QA Leaders Can Calculate True ROI of AI-Powered Test Automation

A new framework aimed at QA leaders argues that traditional ROI formulas designed for scripted automation tools like Selenium fail to capture the full value of modern AI-powered testing platforms. The guide highlights that AI-native testing introduces distinct cost and benefit structures, including self-healing scripts, agentic test generation, and intelligent failure triage, which older calculations overlook. One key gap identified is maintenance: industry data suggests 30–40% of automation engineering time is spent maintaining existing scripts, a burden that AI self-healing capabilities can significantly reduce. Unlike conventional automation, which delivers static returns, AI testing systems are said to improve over time by learning from execution history and defect patterns, producing compounding rather than linear value. The framework proposes measuring returns across four categories to help QA teams build business cases that resonate with both finance and engineering leadership.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Vulnerability Exploitation Now Top Attack Vector as Patch Times Worsen in 2026

A record 48,185 CVEs were published in 2025, a 20.6% increase over the prior year, while the median time to patch vulnerabilities rose 34% to 43 days. Verizon's 2026 Data Breach Investigations Report found that vulnerability exploitation surpassed credential abuse to become the most common initial attack method for the first time in the report's 19-year history. To address this, security experts recommend formalizing vulnerability remediation SLAs that set strict, severity-based patching deadlines — with critical flaws requiring fixes within 14 days. CISA's new federal directive, BOD 26-04, moves beyond simple CVSS scoring by requiring agencies to assess asset exposure and other contextual risk factors, with full compliance expected by December 7, 2026. Although the directive applies only to federal civilian agencies, CISA has urged private organizations to adopt similar frameworks, and cyber insurers and auditors are already following suit.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

A Structured Roadmap for Aspiring Software Architects: What to Learn and When

Becoming a software architect requires mastering a broad range of skills beyond writing code, including system design, security, cloud platforms, and DevOps. Experts recommend following a layered learning path that begins with proficiency in one programming language and strong computer science fundamentals before advancing to backend development, API design, and databases. Aspiring architects should then build knowledge of cloud platforms such as AWS, containerization tools like Docker, and CI/CD pipelines to understand how software reaches production. Architecture-specific skills such as system design patterns, application security, monitoring, and technical documentation round out the roadmap. The emphasis is on knowing when to apply each concept rather than memorizing tools or frameworks in isolation.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Why Most Indie Hacker Launches Fail and What to Do Before Launch Day

Most indie hacker product launches fail not because of poor execution on launch day, but due to inadequate preparation in the months beforehand. Building in public is widely recommended but primarily attracts fellow founders rather than actual paying customers, creating misleading momentum. Founders are advised to identify where their real target buyers spend time — niche forums, Slack groups, or industry newsletters — and build an audience there instead. Positioning is highlighted as a critical pre-launch step, framed not as a tagline exercise but as a clear answer to who the product is for and why they should change their current behavior. One B2B founder's sign-ups jumped from four to twenty-three per week simply by rewriting copy to emphasize the cost of inaction rather than listing product features.

0 comments Read more at DEV Community