{"id":50813515,"url":"https://github.com/locize/locize-agents","last_synced_at":"2026-06-13T07:04:45.568Z","repository":{"id":363861136,"uuid":"1263691472","full_name":"locize/locize-agents","owner":"locize","description":"Conventions \u0026 setup for using Locize from AI coding agents (MCP context tools: glossary, style guide, translation memory, screenshots)","archived":false,"fork":false,"pushed_at":"2026-06-10T16:34:19.000Z","size":6,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-06-10T18:14:38.673Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/locize.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null},"funding":{"github":null,"patreon":null,"open_collective":null,"ko_fi":null,"tidelift":null,"community_bridge":null,"liberapay":null,"issuehunt":null,"otechie":null,"custom":"https://locize.com"}},"created_at":"2026-06-09T07:20:15.000Z","updated_at":"2026-06-10T16:34:24.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/locize/locize-agents","commit_stats":null,"previous_names":["locize/locize-agents"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/locize/locize-agents","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/locize%2Flocize-agents","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/locize%2Flocize-agents/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/locize%2Flocize-agents/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/locize%2Flocize-agents/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/locize","download_url":"https://codeload.github.com/locize/locize-agents/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/locize%2Flocize-agents/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34275068,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-13T02:00:06.617Z","response_time":62,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2026-06-13T07:04:44.782Z","updated_at":"2026-06-13T07:04:45.560Z","avatar_url":"https://github.com/locize.png","language":null,"funding_links":["https://locize.com"],"categories":[],"sub_categories":[],"readme":"# Locize for AI agents\n\nConventions and setup for using [Locize](https://www.locize.com) from AI coding agents\n(Claude Code, Cursor, VS Code, and any [MCP](https://modelcontextprotocol.io)-compatible\nclient) — so an agent translates, reviews, and lints terminology to your project's\nstandards, right in your editor.\n\nThis repo is the canonical home of the agent-convention snippets. Drop the one that\nmatches your tool into your project, or copy the shared rules into your existing\n`CLAUDE.md` / `AGENTS.md` / `.cursorrules`.\n\n- [`CLAUDE.md`](./CLAUDE.md) — for Claude Code\n- [`AGENTS.md`](./AGENTS.md) — generic (Codex, and tools that read AGENTS.md)\n- [`.cursorrules`](./.cursorrules) — for Cursor\n\nThey all carry the same rules; pick the file your tool reads.\n\n## Connect the Locize MCP server\n\nThe conventions assume your agent can reach the Locize MCP server:\n\n```bash\n# Claude Code\nclaude mcp add --transport http locize https://mcp.locize.app\n```\n\nFor Claude Desktop, Cursor, VS Code, and other clients, add `https://mcp.locize.app`\nas a custom MCP connector and complete the OAuth flow, or use a\n[Personal Access Token](https://www.locize.com/docs/integration/personal-access-tokens)\n(`read` scope is enough for the context tools below). Full setup:\n\u003chttps://www.locize.com/docs/integration/mcp\u003e.\n\n## The context tools these conventions use\n\nAll read-only, `read` scope:\n\n| Tool | What it gives the agent |\n|------|-------------------------|\n| `get_glossary` | Approved \u0026 forbidden terms per language (`preferred` / `allowed` / `forbidden`) |\n| `get_styleguide` | Tone, formality, audience, usage rules per language |\n| `search_translation_memory` | Exact + fuzzy matches from prior translations |\n| `get_screenshot_context` | The screenshot + region where a key appears |\n\n## What the conventions tell the agent\n\n- **Before translating or reviewing**, fetch context: use each glossary term's\n  `preferred` value (never a `forbidden` one), apply the style guide's tone/formality,\n  and reuse high-score translation-memory matches. Pull screenshot context when a string\n  is ambiguous.\n- **For pull-request terminology lint** (no translation needed): call\n  `get_glossary(languages: \"\u003csource\u003e\")` and flag any new/changed source string that\n  contains a `forbidden` term, suggesting the language's `preferred` term instead.\n- Matching against source text is the agent's job — terms are returned verbatim (case as\n  stored).\n\nSee [CLAUDE.md](./CLAUDE.md) for the full rule text.\n\n## Localizing an app end-to-end\n\nThe conventions above assume an existing Locize project. To take an app from hardcoded\nstrings to fully localized (detect → instrument → extract → connect to Locize →\nAI-translate → download), the canonical agent flow ships with\n[i18next-cli](https://github.com/i18next/i18next-cli):\n\n```bash\nnpx i18next-cli localize --print-agent-prompt\n```\n\nPaste the printed prompt into your agent, or let it run the command itself. The prompt is\nversion-matched to the installed CLI — regenerate it from the command rather than keeping\na copy in your conventions, so it never drifts from what the CLI actually does. (Without\nan agent, `npx i18next-cli localize` runs the same flow as one interactive command.)\n\nMore: [launch post](https://www.locize.com/blog/i18next-cli-localize?from=locize-agents_readme__localize)\n· [CLI docs](https://www.locize.com/docs/integration/cli?from=locize-agents_readme__localize).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flocize%2Flocize-agents","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flocize%2Flocize-agents","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flocize%2Flocize-agents/lists"}