SShortSingh.
Back to feed

drift: Open-Source Mandate Enforcement for AI Agents Spending Money

0
·1 views

A developer has released drift, an open-source tool that enforces spending mandates on AI agents by checking limits on payee, amount, currency, time window, and audience before any action runs. The rules deny by default and are exposed through MCP tools built on the Python standard library. The project was motivated by an alleged gap in Google's AP2 Python SDK, where the audience check on terminal hops is skipped for certain key-bound tokens, allowing a token minted for one audience to be accepted by another. The author says this flaw lets a token issued for one merchant be accepted by a different party. Code, a write-up, and a reproducer are published on GitHub and the project site.

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 ·

TrustIntern: Student-built tool flags internship scams with evidence-backed risk scores

A developer built TrustIntern, a tool that helps students check internship offers for signs of fraud. Users paste in a recruiter's email, company details and the offer message, and can attach the offer letter as a PDF or screenshot. The tool runs fixed checks such as email domain matching, domain age, and the wording of the message, alongside research agents that look into the company and public complaints. Each risk score is broken down into specific, traceable pieces of evidence, and reports state they are heuristic indicators rather than proof of fraud. The student version is free, while colleges, job boards and paid postmortem reports for scam victims are planned revenue sources.

0
ProgrammingDEV Community ·

DeepSeek Releases Cheaper, Faster Open-Weight Coding Model V4.1-Flash-1017

DeepSeek released DeepSeek-V4.1-Flash-1017 on October 17, 2026, a minor update to its V4.1-Flash model aimed at agentic coding work. According to vendor-reported benchmarks, the model posts higher scores than its predecessor, completes tasks faster, and produces shorter outputs, while offering a 1M-token context window and MIT-licensed open weights. It leads several agentic coding benchmarks but trails frontier models like Claude Opus 5 on harder reasoning tests, and no independent evaluation exists yet. DeepSeek's API pricing is substantially lower than closed frontier models, with cache hits costing far less than cache misses and off-peak rates at half the peak level. The company also opened its Search & Fetch APIs to all users at $0.75 per 1,000 successful calls.

0
ProgrammingDEV Community ·

NexTrace Uses FalkorDB Graph Database to Detect and Quarantine Cloud Attack Paths

A developer built NexTrace, an AI security agent for the FalkorDB Hackathon, to address cloud breaches where attackers chain small flaws through internal permissions. Traditional tools only send alerts without revealing attack routes or fixing them, and many AI chatbots can only offer written advice. NexTrace models cloud services and their connections as a graph in FalkorDB, allowing it to map the blast radius and find the shortest attack path to sensitive data. The agent can then take direct action by quarantining the vulnerable service and severing risky connections, before verifying that no path to the target remains.

0
ProgrammingDEV Community ·

First-Year CS Student Shares Why He Chose Computer Science and His Goals

Allen, a first-year Bachelor of Science in Computer Science student, explains in a personal blog post why he chose the field. He wants to understand how programs work and how technology fits into everyday life, and he had little programming experience before enrolling. He is currently learning C++, covering variables, data types, input and output, and conditional logic, and he says practice is helping him work through errors. Looking ahead, he hopes to build useful, easy-to-use programs and applications and to gain experience through coursework and personal projects. He also encourages fellow beginners to treat mistakes as part of learning.