SShortSingh.
0
ProgrammingDEV Community ·

Dental Clinic Sees 247% More Calls in 90 Days Through Technical SEO Changes

A dental clinic rose from sixth to the top three in Google's Map Pack, and its total tracked calls grew 247% over 90 days. The author attributes much of the gain to technical work, including adding LocalBusiness structured data and validating it. Other fixes included making the phone number a tappable link on every service page and speeding up high-intent service pages on mobile. Each key treatment query was given its own dedicated page, and business details were kept consistent across the profile, site, and schema. Email recall and follow-up campaigns were also credited with bringing dormant patients back for recurring visits.

0
ProgrammingDEV Community ·

Microsoft's AutoGen Research Group Disbanded; AutoGen No Longer Receives Security Fixes

John Langford, who helped establish Microsoft Research's AI Frontiers group, announced on November 6 that the group is over after its members were reorganized into Microsoft AI. The group's work included the AutoGen orchestration framework and the Fara computer use agent. Microsoft had already stated on October 16 in the AutoGen repository README that it provides no bug fixes, security fixes, or support for the project, and advised against production use. The repository is not flagged as archived on GitHub, so dependency audits may not surface the warning. Microsoft directs users to its Microsoft Agent Framework as the successor to AutoGen.

0
ProgrammingDEV Community ·

osFoundry enforces Claude Code tool allowlists on its MCP server, not through CLI flags

osFoundry's desktop app passes an MCP configuration to Claude Code when it runs as a worker agent, binding the bridge to localhost on an OS-assigned port with a per-run token. The author explains that the agent runs with bypassPermissions, which makes Claude Code ignore the --allowedTools flag, so the flag offers no real protection. Instead, the server filters the tool list and checks every tools/call against the run's allowed tools, since the model could still attempt calls to unlisted tools via hallucination or prompt injection. The piece also notes that notifications must return a 202 with an empty body to avoid breaking some MCP clients such as Codex. A rebuilt Python version of the bridge is described as a way for readers to test the behavior themselves.

0
ProgrammingDEV Community ·

Axios CVE-2026-42037: Incident response steps for multipart header injection

CVE-2026-42037 is a CWE-93 flaw in Axios multipart bodies that affects versions from 1.0.0 up to but not including 1.15.1, with a CVSS 3.1 score of 5.3. The guidance urges responders to check whether an affected version was deployed, whether untrusted Blob metadata reached it, and whether a downstream parser acted on an injected header. Teams should gather request metadata and build details, then reproduce the issue locally against a vulnerable and a patched version without sending crafted requests to third parties or production systems. Logs should be reviewed for unexpected part headers, and each signal should be judged by whether it led to a consequential action rather than treated as proof of exploitation. The recommended remediation is upgrading to Axios 1.15.1 or later, with interim input validation at the upload boundary if immediate patching is not possible.

0
ProgrammingDEV Community ·

Game studio redraws 82 map sprites with AI and human review

Between 28 and 28 September 2026, the developer of Myriad Immortal Sect replaced 82 pieces of map art and added art for 14 objects that previously had none. An image model running on a local machine generated the new sprites, while the selection of targets came from surveys of visual inconsistencies. Frost-country species without art had exposed a bug where the engine displayed a protective artifact frame instead of the intended fallback, which was fixed in code and by adding art. The developer reported prompt lessons, such as palette words leaking into non-building subjects and style references copying content, and built a side-by-side review page where candidates were judged in context with their neighbours. The review showed that many first picks were not the best candidate, and a second round confirmed the redraws of several rows.

0
ProgrammingDEV Community ·

Evenly, a shared money app, learns household splits from corrections stored on Walrus

Evenly is a shared expense app for couples and roommates that lets users log purchases in plain language and proposes how to split them. The app records every purchase, accepted split and correction as a memory on Walrus Memory, using its recall to make later proposals, and its Why panel cites the supporting memories. Settlements are computed in USDC on Sui, with payments made from members' own wallets and detected automatically through Tatum. In tests with automated agent households on Walrus mainnet, the system learned hidden split rules after corrections on the first day and marked all six settlements as settled once real payments were made.

0
IndiaTimes of India ·

Trump Offers UFC Chief Dana White a White House Press Secretary Role

President Donald Trump publicly urged UFC president Dana White to take on the job of White House press secretary. The remark was made in a lighthearted manner, with Trump asking White to accept the position. The offer highlights the close relationship between Trump and the mixed martial arts leader. The article's full details on the setting and any reply from White were not available in the provided text.

0
IndiaTimes of India ·

Speeding Thar Kills Scooter Rider in Kanpur, Drags Body 20 Metres Before Fleeing

A man died after a speeding Mahindra Thar struck his scooter in Kalyanpur, Kanpur. The SUV reportedly dragged the rider for close to 20 metres, causing heavy damage to the scene. Eyewitnesses said the vehicle's occupants, two women and a man, fled the spot in another vehicle after the crash. Police have taken the body into custody and are examining nearby CCTV footage to trace those involved.

0
IndiaTimes of India ·

Neighbour Kills Singapore-Based IT Executive After Parking Dispute in Faridabad

A neighbour allegedly fired 15 rounds and killed a Singapore-based IT vice president in Faridabad following a dispute over parking. Reports indicate the two families had also quarrelled over a pet dog before the confrontation escalated. The executive had returned to India to attend a nephew's wedding when the incident occurred. The article provided only limited details beyond these points, so further facts about the investigation or the accused's arrest remain unconfirmed in this source.

0
Crypto & Web3CoinDesk ·

Binance tells users in several European countries to withdraw crypto before accounts close by year end

Binance has notified users in some European countries that they must move their crypto holdings out before new fees take effect. Affected accounts are reportedly scheduled to close on December 31. Users in Spain, France, and Poland say they have received similar notices. The source article provides limited detail beyond these points, and the reasons behind the closures were not specified in the text provided.

0
BusinessEconomic Times ·

Trump Reportedly Floats UFC President Dana White for White House Press Secretary Role

President Donald Trump has reportedly suggested that Dana White, the president of the Ultimate Fighting Championship, could take on the role of White House press secretary. The idea was raised as a potential appointment for the post responsible for briefing the press on the administration's policies. White, a longtime Trump associate and frequent presence at his events, has not been widely reported as having accepted or formally considered the position. The suggestion comes amid ongoing changes in the White House communications team. Details of the discussions remain limited in the source report.

0
ProgrammingDEV Community ·

Fault-injection drill tests offline fallback in local-first apps by cutting the network mid-request

The article argues that a local-first app only proves its offline behavior when the network fails during an in-flight request, not between requests. It recommends a stdlib-only Python HTTP stub that can simulate four failure modes: connection reset, stalled body, truncated response, and normal success. Running the client against each mode is meant to expose bugs such as a truncated response being parsed and reported as success. The author stresses writing explicit assertions before testing, including a local-write latency budget, a check that queued items are not lost, and a check against duplicate sends on retry. Generating an idempotency key before the first attempt is presented as the safeguard against double-writes in reset scenarios.

0
ProgrammingDEV Community ·

Tabs in One: un manager Chrome gratuit care regrupează filele pe o singură coloană și eliberează RAM

Dezvoltatorul Tabs in One, o extensie gratuită pentru Chrome, a lansat un instrument care mută toate filele deschise pe o singură pagină cu o coloană lizibilă. Motivația este că bara orizontală de file devine greu de folosit când sunt deschise peste zece file, iar filele consumă multă memorie. Închiderea filelor originale după mutare eliberează imediat RAM-ul, iar utilizatorii pot restaura oricare sau toate filele cu un singur clic. Extensia oferă backup-uri automate, export și import în format JSON sau HTML și nu necesită cont sau abonament. Ea nu sincronizează sesiunile între dispozitive, iar codul sursă este deschis pentru verificare.

0
ProgrammingDEV Community ·

Developer builds SHADOW, a local platform for supervised AI agents on Windows

A computer science junior describes building SHADOW, a platform for deploying AI agents that run on the user's own PC with their own API keys. Each agent has its own chat, model, permissions, and schedule, and works from a goal by first producing a plan the user must approve. Agents can only access folders and run commands the user explicitly grants, must show risky commands before executing, and cannot control the screen without the user starting it. The author says the design grew out of frustration with manually coordinating separate chat, documentation, and terminal tools. SHADOW is free on the Microsoft Store for Windows 11 and requires no account.

0
ProgrammingDEV Community ·

Keeping a human approval gate between LLM agent proposals and system writes

An AI agent can prepare a consequential change without being permitted to carry it out, which lets organizations automate work while keeping a person accountable. The article uses a supplier onboarding example, where the agent validates data and drafts an ERP record but cannot create the supplier or release payments. It recommends a state machine with a persisted, versioned proposal that reviewers approve exactly, along with separate execution endpoints that accept only valid approval tokens. Approval records should capture identity, payload hash, decision and expiry, and executors need idempotency keys and reconciliation. The author suggests testing rejection, expiry, modification, duplicate and outage paths, and piloting one low-volume workflow with measured outcomes.

0
ProgrammingDEV Community ·

Spectre.Console fixes markup crash when interpolating non-string values with brackets

Spectre.Console's interpolated markup methods only escaped square brackets in string values, so other objects whose ToString() output contained brackets, such as List`1[System.Int32], could trigger a markup parsing exception. The issue was reported as #1763 by @yaakov-h, and a fix was merged on 12 May 2026 as #2118. The new approach converts non-formattable objects to strings and escapes them before formatting, while leaving numbers, dates and other IFormattable values untouched. The change shipped in version 0.56.0, and the author notes that date formats containing brackets still behave differently by design.

0
ProgrammingDEV Community ·

Review Agent-Generated PRs by Rollback Cost Before Reading the Logic

The author argues that the most dangerous change in an agent-generated pull request is one that cannot be reverted, such as a dropped column, rather than a logically wrong line. Because agents usually produce plausible code that passes a quick read, the failures that matter tend to be structural, including migrations, broken imports, unread lockfile bumps, and skipped tests. The proposed review order is to classify each changed path by rollback cost into irreversible, mutating, ordinary, or additive buckets, split irreversible changes into their own PRs, and test the rollback before reading logic. The post includes a git-only Python classifier and a decision table, which the author notes has only been tested against synthetic repositories and should be validated before real use.

0
ProgrammingDEV Community ·

How App Store search rankings favor new apps, based on 350,000 results across 87 countries

A developer analyzed Apple's most popular search lists, covering up to 500 searches per category in 87 countries, and collected about 350,000 App Store results. Among new apps released in the past six months, roughly 42,000 reached a top-10 position somewhere, and most of these had fewer than 10 ratings. The findings indicate that placing a popular search term at the start of an app's name is one of the strongest ranking factors, improving top-10 odds substantially compared with brand-first names. Other factors include competition from rival apps using the same name and the number of ratings an app has. The guide also explains how the subtitle, keyword field, and description each contribute differently to visibility.

0
ProgrammingDEV Community ·

DEV Community lab proposes measuring routing tasks between free model endpoint and self-hosted server

A DEV Community article presents a 90-minute workshop for teams to decide which execution path handles each class of task, based on their own measurements rather than public leaderboard scores. The lab uses a frozen prompt set, a JSONL evidence log, and a twelve-row ledger covering three task kinds across four execution paths. Only schema-valid responses count as successes, so fast malformed outputs are treated as failures. The workshop culminates in a routing policy table that participants can rerun after endpoint changes. A reference harness script is included, though it has not been run against a live endpoint and must be adapted to the endpoint in use.

← NewerPage 475 of 7352Older →