Scion

gitai is an open-source, MIT-licensed version-control system designed for AI coding agents. It moves repository operations onto a server: clients fetch individual objects and sparse trees, while the server handles history queries, blame, path history, ancestry, code search, and derived-data caching. Changes land by rebasing onto the current branch tip and using compare-and-swap rather than a merge queue; concurrent lands can be batched, and accepted reference updates are recorded in an ordered changeset log. Conflicts can be resolved through a versioned, vendor-neutral protocol using an LLM provider, a deterministic tool, or no resolver. The server uses normal Git repositories underneath, with SQLite or PostgreSQL for changesets and references and file or cloud object storage for content-addressed packs. It includes a graft CLI and Go SDK, optional cache daemons, agent tokens and branch policies, semantic review rules, post-landing verification and reversion, and agents that monitor changesets. The repository describes it as an independent implementation inspired by Anthropic's internally developed Scion design, not affiliated with or endorsed by Anthropic.

Antique is Anthropic’s in-house code-review surface, built and shipped alongside Scion. It applies semantic constraints to approvals, distinguishing comment-only changes from sensitive logical changes, and supports an Antique change-request workflow in place of GitHub pull requests.

Sourcegraph is a code-understanding and code-evolution platform for indexing, searching, and overseeing large codebases across monorepos and multiple repositories. It provides exact, deterministic code search, SCIP-powered context through an MCP server, natural-language Deep Search with citations, code oversight and insights, monitoring, living documentation, and tools for tracking system-wide impact as code changes. Its Agentic Batch Changes capability applies an AI agent to migrations, modernization, and remediation across repositories, while retaining control over the resulting changes. The platform is designed to give coding agents broader codebase context so they can find cross-cutting dependencies and update affected layers rather than relying on isolated file searches. Sourcegraph states that it serves enterprise engineering teams, supports large monorepos and multi-repository architectures, and offers SOC 2 Type II and ISO 27001 compliance with zero data retention.

Ant CLI is a command-line tool from Anthropic for managing Claude-based agents. It lets developers define long-term agent resources in YAML within a codebase and synchronize or update those definitions from CI/CD pipelines in a Terraform-style workflow. The workshop describes its use for versioning agent configurations and supporting tenant isolation across workspaces, organizations, or session metadata.

Agentic Batch Changes is a Sourcegraph product for coordinating code changes across hundreds or thousands of repositories, or across large monorepos. It is built on Sourcegraph's Batch Changes and Deep Search technologies: an engineer describes a change in plain language, the system scopes it across an indexed codebase, builds a plan, tests the change in one repository, and rolls it out while adapting to repository differences. For each part of the plan, it chooses between generating a deterministic script and assigning a coding agent with repository context. It opens pull requests, responds to CI failures, tracks their status through merge, displays generated diffs, and pauses for human decisions when needed. Sourcegraph states that the product is available to Sourcegraph Cloud customers and uses outcome-based pricing based on changesets merged rather than tokens, seats, or attempts.