--- title: Detect vulnerabilities in LLM applications with... description: Datadog Code Security's AI-native SAST detects vulnerabilities specific to the OWASP Top 10 for LLM Applications, such as prompt injection, excessive agency,... canonical: https://daily.dev/posts/detect-vulnerabilities-in-llm-applications-with-datadog-s-ai-native-sast-v8pbj9hla twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Detect vulnerabilities in LLM applications with Datadog’s AI-native SAST | daily.dev og:description: Datadog Code Security's AI-native SAST detects vulnerabilities specific to the OWASP Top 10 for LLM Applications, such as prompt injection, excessive agency,... og:url: https://daily.dev/posts/detect-vulnerabilities-in-llm-applications-with-datadog-s-ai-native-sast-v8pbj9hla og:image: https://api.daily.dev/og/posts/v8pBj9Hla.png og:image:alt: Detect vulnerabilities in LLM applications with Datadog’s AI-native SAST og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Detect vulnerabilities in LLM applications with Datadog’s AI-native SAST **[Datadog](https://daily.dev/sources/datadog)** · 5 min read · 0 upvotes · 0 comments ## Summary Datadog Code Security's AI-native SAST detects vulnerabilities specific to the OWASP Top 10 for LLM Applications, such as prompt injection, excessive agency, and hidden context exposure. Instead of relying purely on pattern matching, it uses taint analysis and control flow analysis to trace how untrusted input reaches LLM calls and what capabilities an agent is granted, then independently verifies candidate findings. Coverage spans Python, Go, Java, C#, TypeScript, and JavaScript, with results surfaced as PR comments and CI checks via PR Gates. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How does static analysis detect prompt injection vulnerabilities in LLM application code? Taint analysis traces the flow of untrusted input, such as a user-supplied query parameter, to see whether it reaches an LLM call without sanitization. If a variable like a raw query string is concatenated directly into a prompt sent to an LLM, that path is flagged as a prompt injection sink, since the model could treat attacker-controlled data as an instruction rather than content. _Teams shipping LLM features can track new SAST techniques for prompt injection risks via daily.dev._ ### What is excessive agency in LLM applications and how can it be detected before deployment? Excessive agency (OWASP LLM03) happens when an LLM-powered agent is granted consequential capabilities, like shell access, file system access, or external API calls, without proper authorization checks. Control flow analysis examines all code paths that initialize the agent's toolset, flagging any path that grants unrestricted access regardless of user permissions, even across multiple callers or conditional branches. _Developers wiring up agent tool permissions can follow LLM security detection techniques on daily.dev._ ### Which programming languages does Datadog's AI-native SAST support for LLM vulnerability detection? Coverage for the OWASP Top 10 for LLM Applications spans Python, Go, Java, C#, TypeScript, and JavaScript. Detection combines taint analysis for prompt injection, control flow analysis for excessive agency, and pattern matching for hidden context exposure, such as system prompts accidentally logged via calls like logger.info(). _Teams evaluating security tooling for polyglot LLM codebases can compare options on daily.dev._ ## Similar posts on daily.dev - [Introducing our open source AI-native SAST](https://daily.dev/posts/introducing-our-open-source-ai-native-sast-ufkhzq6e7) · Datadog · 16 upvotes · 0 comments - [Using LLMs to filter out false positives from static code analysis](https://daily.dev/posts/using-llms-to-filter-out-false-positives-from-static-code-analysis-nozql7azs) · Datadog · 0 upvotes · 0 comments - [AI Is Building Your Attack Surface. Are You Testing It?](https://daily.dev/posts/ai-is-building-your-attack-surface-are-you-testing-it--gue53d6fr) · Snyk · 0 upvotes · 0 comments - [AI SAST: Code Security for the Agentic SDLC](https://daily.dev/posts/ai-sast-code-security-for-the-agentic-sdlc-sxvc4pirt) · SD Times · 0 upvotes · 0 comments - [When to use SAST versus an LLM security scanner](https://daily.dev/posts/when-to-use-sast-versus-an-llm-security-scanner-g0xclaxzt) · GitLab · 0 upvotes · 0 comments --- Tags: [#security](https://daily.dev/tags/security), [#devops](https://daily.dev/tags/devops), [#prompt-injection](https://daily.dev/tags/prompt-injection) [View this post on daily.dev](https://daily.dev/posts/detect-vulnerabilities-in-llm-applications-with-datadog-s-ai-native-sast-v8pbj9hla) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Detect vulnerabilities in LLM applications with Datadog’s AI-native SAST","url":"https://daily.dev/posts/detect-vulnerabilities-in-llm-applications-with-datadog-s-ai-native-sast-v8pbj9hla","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/detect-vulnerabilities-in-llm-applications-with-datadog-s-ai-native-sast-v8pbj9hla"},"datePublished":"2026-08-20T15:03:52.809Z","dateModified":"2026-09-13T20:46:32.954Z","description":"Datadog Code Security's AI-native SAST detects vulnerabilities specific to the OWASP Top 10 for LLM Applications, such as prompt injection, excessive agency,...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b6da4ce1a59b4dd74232c958c4a7a6bf?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b6da4ce1a59b4dd74232c958c4a7a6bf?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Datadog","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Datadog","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/8f9d7e3e4bd64bd7b9db9b2ea83e0c6a","url":"https://daily.dev/sources/datadog"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/detect-vulnerabilities-in-llm-applications-with-datadog-s-ai-native-sast-v8pbj9hla","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"security,devops,prompt-injection","timeRequired":"PT5M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Datadog","item":"https://daily.dev/sources/datadog"},{"@type":"ListItem","position":3,"name":"Detect vulnerabilities in LLM applications with Datadog’s AI-native SAST"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/detect-vulnerabilities-in-llm-applications-with-datadog-s-ai-native-sast-v8pbj9hla#faq","mainEntity":[{"@type":"Question","name":"How does static analysis detect prompt injection vulnerabilities in LLM application code?","acceptedAnswer":{"@type":"Answer","text":"Taint analysis traces the flow of untrusted input, such as a user-supplied query parameter, to see whether it reaches an LLM call without sanitization. If a variable like a raw query string is concatenated directly into a prompt sent to an LLM, that path is flagged as a prompt injection sink, since the model could treat attacker-controlled data as an instruction rather than content. Teams shipping LLM features can track new SAST techniques for prompt injection risks via daily.dev."}},{"@type":"Question","name":"What is excessive agency in LLM applications and how can it be detected before deployment?","acceptedAnswer":{"@type":"Answer","text":"Excessive agency (OWASP LLM03) happens when an LLM-powered agent is granted consequential capabilities, like shell access, file system access, or external API calls, without proper authorization checks. Control flow analysis examines all code paths that initialize the agent's toolset, flagging any path that grants unrestricted access regardless of user permissions, even across multiple callers or conditional branches. Developers wiring up agent tool permissions can follow LLM security detection techniques on daily.dev."}},{"@type":"Question","name":"Which programming languages does Datadog's AI-native SAST support for LLM vulnerability detection?","acceptedAnswer":{"@type":"Answer","text":"Coverage for the OWASP Top 10 for LLM Applications spans Python, Go, Java, C#, TypeScript, and JavaScript. Detection combines taint analysis for prompt injection, control flow analysis for excessive agency, and pattern matching for hidden context exposure, such as system prompts accidentally logged via calls like logger.info(). Teams evaluating security tooling for polyglot LLM codebases can compare options on daily.dev."}}]} ```