The latest articles on DEV Community by Ganesh Joshi (@ganeshjoshi). https://dev.to/ganeshjoshi https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3812945%2F5c98e077-4bc4-4521-8bf1-006bca4726f1.png https://dev.to/ganeshjoshi en Ganesh Joshi Sat, 20 Jun 2026 11:21:50 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-kai https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-kai <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Fri, 19 Jun 2026 12:51:52 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-2369 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-2369 <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Thu, 18 Jun 2026 12:29:30 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-18hm https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-18hm <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Wed, 17 Jun 2026 12:54:27 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-4cm6 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-4cm6 <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Tue, 16 Jun 2026 14:21:46 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-56kj https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-56kj <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Mon, 15 Jun 2026 15:20:18 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-k59 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-k59 <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Sun, 14 Jun 2026 11:33:43 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-4df2 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-4df2 <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Sat, 13 Jun 2026 11:20:33 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-mge https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-mge <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Fri, 12 Jun 2026 12:23:57 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-5h51 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-5h51 <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Thu, 11 Jun 2026 12:50:14 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-4hnf https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-4hnf <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Wed, 10 Jun 2026 12:29:03 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-12cb https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-12cb <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai Ganesh Joshi Tue, 09 Jun 2026 12:09:35 +0000 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-3166 https://dev.to/ganeshjoshi/anthropic-api-claude-tool-use-and-structured-outputs-in-apps-3166 <p><em>This post was created with AI assistance and reviewed for accuracy before publishing.</em></p> <p>Anthropic documents the <strong>Messages API</strong>, <strong>models</strong>, and <strong>tool use</strong> for Claude at <a href="https://docs.anthropic.com/en/api" rel="noopener noreferrer">Anthropic API documentation</a>. Model IDs, pricing, and capabilities are versioned; copy model strings from the console or docs, not from old blog posts.</p> <h2> Tools </h2> <p>Tool calling lets Claude request structured actions (HTTP, DB, internal functions) that your server executes. Validate arguments with schemas before execution.</p> <h2> Structured workflows </h2> <p>When you need machine-readable output, combine tool use with strict validation in your own code. Treat model output as untrusted until parsed.</p> <h2> Practical takeaway </h2> <p>Rotate API keys. Log request IDs for support. Monitor token usage against budgets.</p> anthropic claude api ai