GitHits
Show HN: GitHits Public Beta 0.9
Last verified:
What is GitHits?
GitHits is an AI-native code search engine that gives AI coding agents version-aware access to real open-source code, documentation, dependency context, and package metadata. It builds a version-aware index of open-source code, package internals, documentation, dependency graphs, vulnerabilities, changelogs, and package upgrade changes so agents can retrieve and navigate the actual code running in their stack.
The tool searches open-source repositories and distills implementation guidance from existing code instead of returning long lists of search results. It analyzes code, issues, discussions, pull requests, and repository context to produce examples relevant to the problem being investigated. This gives developers and AI coding agents a concrete starting point based on how similar problems were solved in open source.
GitHits is designed for AI coding agents and developers working with tools like Claude Code, Cursor, or Copilot who encounter problems models cannot resolve. It particularly helps engineers building in languages like Go, Rust, Kotlin, or C++ where documentation is thinner and training-data coverage is weaker. The CLI tool is currently in free beta and supports Python, TypeScript, and JavaScript.
Key capabilities include code example search (get_example), code navigation (search, code_grep, code_files, code_read), documentation access (docs_list, docs_read), and package inspection (pkg_info, pkg_deps, pkg_vulns, pkg_changelog). The tool covers 800+ languages for examples and 20+ languages for code navigation.
GitHits pricing
Pricing model: Freemium
GitHits is currently offering a free beta version of the CLI tool available on Product Hunt. The company secured $1.75M in pre-seed financing and plans a commercial release later in 2026. The free beta includes access to code example search (get_example), code navigation tools (search, code_grep, code_files, code_read), documentation access (docs_list, docs_read), and package inspection tools (pkg_info, pkg_deps, pkg_vulns, pkg_changelog). No paid plan pricing details are publicly available yet as the commercial version is not launched.
GitHits pros
- Eliminates AI hallucinations by using real code from GitHub
- Provides distilled code examples instead of raw search results
- Searches all public GitHub repositories, not just local codebase
- Version-aware indexing pinned to immutable commits
- Supports 800+ languages for code example search
- Supports 20+ languages for code navigation including Go, Rust, C++
- Accesses 11 package registries (npm, pypi, hex, crates, etc.)
- No cloning required for code navigation
- Provides dependency graph analysis with transitive resolution
- Includes vulnerability checking with severity filtering
- Accesses both hosted and repo-backed package documentation
- Free beta CLI tool available on Product Hunt
- GitHub authentication for secure access without private repo access
- Guardrails against prompt injection and malicious content
- Does not use customer data to train foundational AI models
- Configurable license filtering (excludes copyleft by default)
- Exact line-range reads from source files
- Deterministic text grep scoped by path/globs/extensions
- Tracks repository activity, adoption, maintenance, and licensing
- Shortens path from uncertainty to working implementation
GitHits cons
- Currently in beta with potential changes to tool names and formats
- Limited to public open-source repositories only
- No access to private repositories or code
- Commercial release not yet available (planned later this year)
- Only 20+ languages fully supported for code navigation
- May have re-indexing risk for moving references like branches
- Requires GitHub authentication to use
- Developers still responsible for reviewing generated code before shipping
- Pre-seed stage company with limited long-term guarantees
- CLI tool setup required via npx githits init
Frequently asked questions about GitHits
What is GitHits and what problem does it solve?
GitHits gives AI coding agents version-aware access to real open-source examples, dependency source code, docs, and package metadata so they can retrieve and navigate the actual code running in your stack. It solves the problem that AI coding agents can read your local repository but the open-source stack underneath (frameworks, SDKs, APIs, dependency internals) stays invisible, leading to retries, fragility, hallucinations, and token burn.
How does GitHits work with AI coding agents?
GitHits complements coding agents that already have access to your local codebase through an MCP server. Agents use GitHits when they need information not available in your local codebase, such as researching an integration, understanding dependency internals, investigating version-specific behavior, or finding how similar problems are solved. Some tools invoke GitHits automatically when needed, while others may require explicit instruction.
Does GitHits access my private repositories?
No. GitHits does not access, index, or store your private source code. It uses GitHub authentication to power code example search and metadata lookups against public open-source repositories only. Connecting GitHub does not give GitHits access to private code, and GitHits does not index, search, or access private repositories.
What languages does GitHits support?
GitHits supports 800+ languages for code example search (get_example). For code navigation tools (search, code_grep, code_files, code_read), it supports 20+ languages: Bash, C#, C++, CSS, Dart, Elixir, Erlang, Go, Java, JavaScript, Kotlin, Lua, Markdown, PHP, Proto, Python, R, Ruby, Rust, Scala, SCSS, Swift, TypeScript, and Zig. The CLI currently supports Python, TypeScript, and JavaScript.
What package registries does GitHits support?
GitHits supports 11 package registries for package inspection tools: npm, pypi, hex, crates, vcpkg, zig, nuget, maven, packagist, rubygems, and go. This enables tools like pkg_info (package overview), pkg_deps (dependency graph analysis), pkg_vulns (vulnerability checks), and pkg_changelog (release notes).
How does GitHits handle version awareness and indexing?
GitHits builds an index by fetching a specific commit for each repository and extracting files, symbols, imports, call relationships, and documentation into a code graph. For pinned versions (package versions like 0.2.5 or specific commits), the indexed content is stable and reproducible. The index is pinned to immutable commits and updates as repositories and packages change. Moving references like branches or HEAD intentionally track the latest commit.
Is GitHits safe regarding prompt injection and malicious content?
GitHits includes guardrails designed to reduce prompt injection and other malicious content risks in both their search infrastructure and MCP tools. Because agents retrieve structured code, documentation, and source references rather than browsing arbitrary web pages, GitHits generally provides a more controlled source of context than web search. However, as with any third-party source, developers remain responsible for reviewing generated code before shipping.
How does GitHits handle license filtering?
GitHits can be configured to exclude repositories with specific licenses. By default, example generation runs in strict mode, which excludes repositories with copyleft licenses and repositories that do not declare a license. Users can relax these restrictions or disable license filtering entirely for broader coverage. GitHits also exposes license information through package inspection tools, allowing agents to inspect licenses before recommending or using a dependency.
How do I get started with GitHits?
To get started, run 'npx githits@latest init'. GitHits detects your coding tool, signs you in with GitHub, and configures the MCP server automatically. The free beta CLI tool is available on Product Hunt. After setup, agents use GitHits tools like get_example for code examples, search/code_grep/code_read for code navigation, and pkg_info/pkg_vulns for package inspection.
Does GitHits use my code to train AI models?
No. GitHits does not use customer code, prompts, outputs, or personal data to train foundational AI models. Customer data is processed only to provide the service and remains governed by their Privacy Policy and Data Processing Agreement. GitHits is designed to complement coding agents without requiring access to your private repositories.