# Amplifying: product and company guide This file combines the public product and company pages and the research directory. It is not a full dump of the website. Read individual research articles and data indexes through [llms.txt](https://amplifying.ai/llms.txt). Canonical page: https://amplifying.ai/ Markdown exported: 2026-10-25. This is the export date, not the date of the observations. Research dates and limitations remain in the article. This text captures the page's default reading view. Use the canonical page for interactive charts, filters, and additional views. Product demonstrations, including Beacon, are hypothetical examples, not customer results or measured research findings. # Understand how AI agents [choose](), [build](), and [perform](). Benchmarks and research on the choices agents make, the code they write, and the tasks they complete. [Set up a call]() [Explore the products]() Record what coding agents do. See what to improve. Used by leading **enterprise DevTools and infrastructure companies.** From the researchers behind the original [What Claude Code Chooses]() study. Latest research / October 6, 2026 ## [Dots vs. Muse for Developers]() We gave both agents the same app prompt. Dots published through ChatGPT Sites; Muse built local apps and tried tunnels for public access. Compare the stacks, screenshots and results. [Read the research]() [Browse all research]() [Explore the Coding Agents Index]() Two products ## [See what agents choose.]() [The lineup: compare how coding agents choose between products.]() 01 / Choices, findings, playbooks ### [Benchmarking]() Measure when coding agents choose your product, what they choose instead, and why. Get findings and prioritized playbooks for your team. - Product selection and competitive benchmarks - Sources and evidence behind each finding - Playbooks for docs, positioning, product features, and developer experience [Explore Benchmarking]() ## [Test what helps them succeed.]() [The control room: change the inputs and measure the difference.]() 02 / Changes, tests, results ### [Ground Control]() Make a change to your docs, skills, or tooling. Rerun the same task to see what gets better or worse, then iterate. - Installation, authentication, and real workflows - Recorded runs with inspectable tool calls - Before-and-after tests of docs and tooling [Explore Ground Control]() More from our research ## Tool choice and security. [Browse all research]() Our original study / February 2026 ### [We pioneered research on what coding agents choose.]() Our original *What Claude Code Chooses* study put numbers behind agents’ developer-tool decisions. Today, we help developer tool teams measure and improve how coding agents find, understand, and use their products. [Read the original study]() Safety / April 2026 ### [What security defaults do agents choose?]() Six development tasks leave the security defaults unspecified. The report checks the resulting code against 33 exploit tests. [Read study]() ## Get research updates New Amplifying research on coding agents, delivered by email. ## See Amplifying in action. Tell us what you’re working on. We’ll show you where Amplifying can help. [Set up a call]() --- [All agent-readable resources](https://amplifying.ai/llms.txt) --- Canonical page: https://amplifying.ai/benchmarking Markdown exported: 2026-10-25. This is the export date, not the date of the observations. Research dates and limitations remain in the article. This text captures the page's default reading view. Use the canonical page for interactive charts, filters, and additional views. Product demonstrations, including Beacon, are hypothetical examples, not customer results or measured research findings. Benchmarking # Know where coding agents choose you. Know what to improve where they don’t. Benchmark your product against alternatives across real coding tasks. Get findings that explain their choices and a prioritized playbook of recommended changes to positioning, features, docs, and tooling. [See a demo]() [Explore the product]() Used by teams at leading enterprise developer tool and infrastructure companies. The lineup: compare how coding agents choose between products. Benchmarks, findings, and playbooks ## See where your product gets chosen. Run real coding agents in real repositories and isolated environments. Compare choices across tasks, agents, and competitors, then inspect the sessions behind the numbers. **Explore Benchmarking** Explore each product area. See the full workflow in a demo. **Amplifying** Dashboard Beacon workspace Benchmark **Error monitoring** Beacon · Benchmarking Beacon / Benchmarking / Overview Product preview Error monitoring / Product selection ### Where coding agents choose Beacon. Compare your product with alternatives across coding agents and development tasks. **58%** Choose Beacon **150** Agent sessions **3** Coding agents | Product | Primary choice | | --- | --- | | Beacon | 58% | | Alternative A | 24% | | Alternative B | 13.3% | | No selection | 4.7% | ## Understand your position across coding agents. Compare primary choices, alternative recommendations, and mentions across the tasks that matter to your market. See which competitors agents select, and where the pattern differs between coding agents. ## Findings explain what agents do and why it matters. See where agents misunderstand your product, overlook a capability, or choose an alternative. Each finding connects the observed behavior to supporting sessions and citations, so your team can see the evidence behind it. ## A playbook of recommended changes. Amplifying recommends specific changes to your positioning, product features, documentation, examples, and agent tooling, then prioritizes them in an action playbook. Each action explains what to change, which finding supports it, and how to check whether it helped. Use [Ground Control]() to compare how agents complete the task before and after a change. ## See Amplifying in action. Tell us what you’re working on. We’ll show you where Amplifying can help. [Set up a call]() --- [All agent-readable resources](https://amplifying.ai/llms.txt) --- Canonical page: https://amplifying.ai/ground-control Markdown exported: 2026-10-25. This is the export date, not the date of the observations. Research dates and limitations remain in the article. This text captures the page's default reading view. Use the canonical page for interactive charts, filters, and additional views. Product demonstrations, including Beacon, are hypothetical examples, not customer results or measured research findings. Ground Control # Test whether coding agents can use what you ship. Change your docs, skills, or tooling. Repeat the same task to see what improves, what breaks, and where agents get stuck. [See a demo]() [Explore the product]() The control room: change the inputs and measure the difference. Results, findings, and recommendations ## Find what gets in the way of a working setup. Explore experiments, requirements, run traces, findings, and actions. See how the product helps your team move from an incomplete workflow to a change worth testing. **Explore Ground Control** Explore each product area. See the full workflow in a demo. **Amplifying** Dashboard Beacon workspace Benchmark **Error monitoring** Beacon · Ground Control Beacon / Ground Control / Overview Product preview Error monitoring / Setup benchmark ### Does the setup work when an agent does it? See how reliably coding agents complete a workflow with your product, and where they need help. **24 / 30** Successful runs **6** Need attention **3** Coding agents | Coding agent | Successful runs | | --- | --- | | Claude Code | 9 / 10 | | Codex | 7 / 10 | | Gemini CLI | 8 / 10 | ## Test the things your team can change. Compare docs, examples, skills, SDKs, CLIs, and MCP tooling. Start with a workflow that matters to your product, such as setup, authentication, or a first successful API call. ## Keep checking as agents change. Rerun the same tasks when you update your product or an agent releases a new version. Compare completion, steps, retries, and requests for human help, with the runs behind each result. ## Measure choice with Benchmarking. Ground Control tests how well agents use your product. [Benchmarking]() measures when they choose it and produces findings and playbooks for positioning, product features, docs, and tooling. ## See Amplifying in action. Tell us what you’re working on. We’ll show you where Amplifying can help. [Set up a call]() --- [All agent-readable resources](https://amplifying.ai/llms.txt) --- Canonical page: https://amplifying.ai/for-vendors Markdown exported: 2026-10-25. This is the export date, not the date of the observations. Research dates and limitations remain in the article. This text captures the page's default reading view. Use the canonical page for interactive charts, filters, and additional views. Product demonstrations, including Beacon, are hypothetical examples, not customer results or measured research findings. For developer tool providers # Improve how agents find, understand, and use your product. See what agents choose and where they get stuck. Make informed changes to your product, docs, examples, and tooling. Two products ## [See what agents choose.]() [The lineup: compare how coding agents choose between products.]() 01 / Choices, findings, playbooks ### [Benchmarking]() Measure when coding agents choose your product, what they choose instead, and why. Get findings and prioritized playbooks for your team. - Product selection and competitive benchmarks - Sources and evidence behind each finding - Playbooks for docs, positioning, product features, and developer experience [Explore Benchmarking]() ## [Test what helps them succeed.]() [The control room: change the inputs and measure the difference.]() 02 / Changes, tests, results ### [Ground Control]() Make a change to your docs, skills, or tooling. Rerun the same task to see what gets better or worse, then iterate. - Installation, authentication, and real workflows - Recorded runs with inspectable tool calls - Before-and-after tests of docs and tooling [Explore Ground Control]() ## See Amplifying in action. Tell us what you’re working on. We’ll show you where Amplifying can help. [Set up a call]() --- [All agent-readable resources](https://amplifying.ai/llms.txt) --- Canonical page: https://amplifying.ai/contact Markdown exported: 2026-10-25. This is the export date, not the date of the observations. Research dates and limitations remain in the article. This text captures the page's default reading view. Use the canonical page for interactive charts, filters, and additional views. Contact # Set up a call. Tell us what you’re working on or what you’d like to know. Share your Calendly or Cal.com link and we’ll book a time. Otherwise, we’ll arrange one by email. ## Your details --- [All agent-readable resources](https://amplifying.ai/llms.txt) --- Canonical page: https://amplifying.ai/research Markdown exported: 2026-10-25. This is the export date, not the date of the observations. Research dates and limitations remain in the article. This text captures the page's default reading view. Use the canonical page for interactive charts, filters, and additional views. # Amplifying research We study how coding agents find, understand, and use developer tools. ## Reference indexes - [Coding Agents Index](https://amplifying.ai/coding-agents): Public GitHub activity, agent comparisons, and trends. - [Coding Models Index](https://amplifying.ai/models): Model releases, API prices, and dated benchmark results. - [Decision Models Index](https://amplifying.ai/decision-models): Typed-decision models, deployment options, and primary sources. ## Published research - [Dots vs. Muse for Developers](https://amplifying.ai/research/dots-vs-muse.md) (2026-10-06): We gave Dots and Muse the same app prompt. Dots published through ChatGPT Sites and Cloudflare; Muse built local servers and tried tunnels. Compare their stacks, screenshots and results. - [What Muse Actually Chooses: Four Different Stacks, Not Fully Deployed](https://amplifying.ai/research/meta-muse-chooses.md) (2026-10-06): Muse built servers, databases and sign-in across four different stacks. Making the apps publicly accessible remained unfinished. - [What Dots Actually Choose: OpenAI’s Agent Already Has a Stack](https://amplifying.ai/research/openai-dots-choose.md) (2026-10-01): Our benchmark shows OpenAI’s Dots defaults to its own platform, powered by Cloudflare, rather than separate third-party services when building apps. - [Six boundaries to check when sandboxing coding agents](https://amplifying.ai/research/agent-sandbox-escapes.md) (2026-09-25): Lessons from building coding-agent benchmarks. Six isolation boundaries, practical checks, and what transcripts alone cannot establish. - [State of the Coding Agent Market](https://amplifying.ai/research/state-of-coding-agents.md) (2026-09): A view of the coding-agent market through attributed public commits and pull requests, with methods and limits attached to the counts. - [What Kimi K3 & GLM-5.2 Actually Choose (China's AI models vs Claude Sonnet 5)](https://amplifying.ai/research/chinese-models-picks.md) (2026-07): The same developer-tool questions, asked in English and Chinese, produce different choices. Compare Kimi, GLM, and Claude across the same repositories. - [What Fable Actually Chooses](https://amplifying.ai/research/claude-code-picks-fable.md) (2026-07): An examination of Fable’s tool choices and the tasks it solves with custom code. The study also tracks which vendors it names as future replacements. - [The Security Decisions Claude Code and Codex Make](https://amplifying.ai/research/ai-security-decisions.md) (2026-04): Claude Code and Codex receive six development tasks with security defaults left unspecified. The report checks the resulting code against 33 exploit tests. - [Claude Code's Leak: Every Hardcoded Vendor and Tool](https://amplifying.ai/research/claude-code-hardcoded-vendors.md) (2026-03-31): An analysis of vendor references across connectors, documentation access, tool presentation, and other integration layers. - [The Tools OpenAI Agreed to Buy](https://amplifying.ai/research/astral-tools.md) (2026-03): A comparison of Ruff, uv, and other Python tools across seven categories and three repositories. Read the selection counts and benchmark conditions. - [What Codex Actually Chooses (vs Claude Code)](https://amplifying.ai/research/codex-vs-claude-code-picks.md) (2026-03): Codex and Claude Code answer the same prompts in the same repositories. Compare their choices across 12 categories, including custom code. - [What Claude Code Actually Chooses](https://amplifying.ai/research/claude-code-picks.md) (2026-02): Open-ended tasks across real repositories, with no tool names in the prompts. Follow which tools are selected and when Claude builds its own solution. --- [All agent-readable resources](https://amplifying.ai/llms.txt)