Skip to content

Coding & building

Make building, testing, shipping, and automation smoother

Explore tools for developers and product builders across code, testing, deployment, automation, and infrastructure. Check where each tool fits, along with its permissions, data handling, and maintenance boundaries.

16 products

PLACES TO START

Products for coding & building

Shown by most recent addition, not by quality or popularity. Check the use and limits before deciding whether to try one.

Filter by current public availability

Showing 16 products. Check each product page for current regions, devices, and account requirements.

AgentLimb
PilotCoding & building

AgentLimb

HOW IT MAY HELP

An AI agent often rereads the DOM, rediscovers controls, and rechecks the route every time it repeats a web task. That exploration is costly for posting, form filling, dashboard work, research, or QA in a real logged-in session. AgentLimb pairs a Chrome extension with a local bridge so later tasks can reuse what was learned.

BEFORE YOU USE IT

The 85%–90% token savings come from maker claims and one task estimate, not a general result; the store has no ratings and the product is still early. The extension uses high-impact permissions including debugger, nativeMessaging, and all-sites access, and can execute short JavaScript in pages, so local does not mean low risk. Code uses Business Source License 1.1 (BSL 1.1), and commercial use must follow the license.

View details
JetBrains Air
PilotCoding & building

JetBrains Air

HOW IT MAY HELP

Multi-agent development often scatters across terminals, chat windows, and temporary branches, making task state, context, and code changes hard to review together. Air organizes that chain with a task list, Plan, terminal, preview, Git worktrees, and diff review.

BEFORE YOU USE IT

Air remains a public preview; models, permissions, execution location, and cost depend on JetBrains AI, BYOK, or external agents, and worktree isolation does not keep code on-device by itself. Agents can edit files, run commands, and consume quota, so review every diff, permission, and network path. Use the live download page for platform availability; stability and privacy remain untested.

View details
Domain Details
CommercialCoding & building

Domain Details

HOW IT MAY HELP

Domain research often means moving among registries, WHOIS/RDAP, DNS, certificates, Wayback, registrar pricing, and several aftermarket venues. Domain Details reduces that fragmentation with a shared search, browser-local history, bulk entry points, and monitoring alerts.

BEFORE YOU USE IT

'All tools run locally' applies only to some browser tools; registration, marketplace, DNS, monitoring, analytics, and error reporting use remote services. Annual price, watched-domain counts, and unlimited-query language differ across pages, while MCP has a 30-per-minute limit. Domain, price, and certificate data are best effort, so verify important results at the source; some links earn affiliate commissions.

View details
Botoi
CommercialCoding & building

Botoi

HOW IT MAY HELP

Developers and AI agents often integrate separate libraries or providers for hashing, DNS, conversion, QR, currency, text, and network checks. Botoi tries to reduce that repeated integration through shared authentication, a JSON contract, an SDK, and an MCP manifest.

BEFORE YOU USE IT

Site, pricing, SDK, and live-manifest numbers differ for tools, APIs, allowances, and versions, so use the live manifest and current pricing. Tools are not all read-only: some create pastes or short URLs and query external URL/DNS services, so review each before agent access. Pastes may persist for 24 hours, short-link retention is unclear, and the API is not entirely account-free or free.

View details
HeimWall
PilotCoding & building

HeimWall

HOW IT MAY HELP

Developers paste `.env` values, logs, customer records, or proprietary code into Cursor, Claude Code, Copilot, and similar tools. Traditional network DLP may not see an app's composer, while vendor dashboards do not provide one cross-tool signal.

BEFORE YOU USE IT

This is early access: version, rule count, classifier, and proxy descriptions differ across pages, with no public checksum or changelog. Individual is described as local, while team mode uploads events and redacted snippets; Investigation Mode, keys, retention, and employee rights need clarification. Privacy and terms remain drafts and SOC 2/HIPAA are incomplete, so buyers should confirm the DPA, price, and contract.

View details
traceAI
Open sourceCoding & building

traceAI

HOW IT MAY HELP

Ordinary HTTP telemetry can show that a request was slow or failed, but not which model call, retrieval result, tool argument, or agent branch caused it. Wiring every framework to a separate observability product creates another data silo and vendor dependency.

BEFORE YOU USE IT

Full prompts, outputs, tool arguments/results, and error context are captured by default, so redaction must be configured before use; error text may still expose sensitive material. Public sources conflict on license and framework counts, so use the current Apache-2.0 repository and compatibility matrix. Hosted cloud, custom collectors, and self-hosting are distinct boundaries, and buyers should request compliance evidence and region details.

View details
LinkMedic
CommercialCoding & building

LinkMedic

HOW IT MAY HELP

Broken links, failed redirects, and server errors accumulate as content and external destinations change. Many site owners do not need a full SEO suite; they need to know what broke, where it appears, when it changed, and what to fix first.

BEFORE YOU USE IT

Public pages conflict on outbound-link and anchor checks, and honoring target-site robots.txt is still listed as future work; do not scan a site without authorization. Privacy information does not explain retention or deletion for target URLs, scan results, and alerts. History access, annual discounts, and the official domain are also inconsistent.

View details
MobileReady
CommercialCoding & building

MobileReady

HOW IT MAY HELP

Responsive review often jumps among browser developer tools, multiple windows, screenshot and annotation apps, and client presentations. Device sizes, page typography and colors, spacing, and full-page captures are difficult to compare in one context.

BEFORE YOU USE IT

The site's Safari, Internet Explorer, and Edge support statements conflict, so rely on the live store before installing. The extension can read page, tab, and navigation data and may request all-sites access. Privacy text does not map store data categories or site analytics clearly, and customer logos lack public sourcing.

View details
Hangar
LiveCoding & building

Hangar

HOW IT MAY HELP

Long-lived local projects mix source, dependencies, build caches, development servers, Git worktrees, and unpushed work that exists only on one machine. Generic cleaners see file size without project meaning, while terminal scripts require developers to remember every directory, launch command, port, and backup state.

BEFORE YOU USE IT

The site exposes no public privacy policy or terms, leaving license validation, site analytics, and support-chat data unclear. The no-telemetry app still checks its license online, while signing, notarization, and recovery lack public installer evidence. Filename-only secret checks cannot replace content scanning, and automatic cleanup should first be tried on a recoverable copy.

View details
TokRepo
LiveCoding & building

TokRepo

HOW IT MAY HELP

Agent skills, system prompts, MCP configurations, and workflows are scattered across repositories, documentation, and chat history. Before copying them, teams lack a consistent way to assess provenance, permissions, dependencies, target tools, write scope, rollback, and reusable install records.

BEFORE YOU USE IT

A hash proves integrity, not publisher identity, licensing, or safety, and public asset metadata lacks a license field. CLI, trust-manifest, and GitHub versions differ, and the latest package cannot be reproduced directly from the public repository. Default telemetry creates a persistent machine identifier and sends limited metadata without exact retention or deletion details.

View details
CodeAgentSwarm
PilotCoding & building

CodeAgentSwarm

HOW IT MAY HELP

Once several coding agents run in parallel, their execution gains can be erased by polling terminals, reconstructing context, finding sessions waiting for approval, and reconciling overlapping edits. Plain terminal panes can display processes side by side but do not understand each agent's task, state, or change scope.

BEFORE YOU USE IT

Code and prompts bypassing the backend does not mean zero data flow: usage events, Sentry crashes, optional accounts, and Swarmi help messages go online. The policy gives no fixed deletion period for analytics, errors, or support messages. In Turbo Mode, Ask behaves as Allow when confirmations are skipped, so actions that must never auto-run need Deny.

View details
specgit
LiveCoding & building

specgit

HOW IT MAY HELP

Product specs are often written in a document tool while code lives in GitHub. Engineers read a linked doc once, implementation moves on, and the two drift apart. Plain Markdown can colocate them, but shifts Git and PR mechanics onto collaborators who may not know that workflow.

BEFORE YOU USE IT

'Docs stay in GitHub' does not mean zero processing: the App needs repository-content and pull-request write access, and collaboration state may persist for up to seven days. AI requests send relevant text and comments to xAI, which may retain them for up to 30 days. Product analytics and operational logs have separate retention and should be reviewed before granting private-repository access.

View details
CartaStudio
PilotWorkCoding & building

CartaStudio

HOW IT MAY HELP

Real decision criteria are scattered across documents, PDFs, web pages, and individual experience. Generic chat and file search struggle to express applicability, exceptions, and rationale. CartaStudio turns that material into typed, scoped, sourced decision graphs stored as Markdown in a user-chosen folder.

BEFORE YOU USE IT

The product remains at 0.5.x and hosted publishing is in closed beta. We have not installed it to test no-telemetry, local-key, or source-bounded-answer claims. Publishing stores account, content, conversation, and usage data, and the site's analytics script conflicts with its privacy text.

View details
nForms
PilotCoding & building

nForms

HOW IT MAY HELP

A basic contact form often needs separate delivery, bot protection, and accessible validation systems. nForms accepts submissions through a form action or one script tag and layers SHA-256 proof-of-work with ARIA error states.

BEFORE YOU USE IT

The ~200 ms result and zero-conversion-loss claim were not reproduced by us, and low-power devices or hostile traffic may behave differently. Injected ARIA cannot replace keyboard, screen-reader, and human accessibility testing. The 99.9% figure is a target; business users should also review processors, retention, and the DPA.

View details
Open Source AI Canvas
Open sourceCreative workCoding & building

Open Source AI Canvas

HOW IT MAY HELP

When experimenting with several image models, prompts, model APIs, outputs, and versions quickly scatter across tools. Developers lose reproducible working states and struggle to switch between OpenAI and Gemini.

BEFORE YOU USE IT

The empty canvas has limited onboarding and prototype-level recovery. Custom API keys and multi-provider support create security, cost, and protocol-change risks, while a front-end-only model limits persistence and collaboration.

View details
Vibe Coding Skills Assessment
LiveLearningCoding & building

Vibe Coding Skills Assessment

HOW IT MAY HELP

People new to AI-assisted development often do not know which tasks to start with or how to judge their current understanding. This product moves a short assessment and explanation to the front of the learning path.

BEFORE YOU USE IT

A short assessment can guide learning but cannot replace review of real project work, code quality, or collaboration. Scoring needs transparent rationale and should not become a permanent label.

View details