Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/griddynamics/rosetta/init-workspace-flow-codegraphgit clone --depth 1 https://github.com/griddynamics/rosettaWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00018 | $0.01163 |
| Opus 5 | $0.00009 | $0.00581 |
| Sonnet 5 | $0.00004 | $0.00233 |
| Haiku 4.5 | $0.00002 | $0.00116 |
Grade A, and why
init-workspace-flow-codegraph scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<init_workspace_flow_codegraph>
<description_and_purpose> Suggests user to install LSPs xor Code graphs only. </description_and_purpose>
<workflow_context>
- Phase 6 of 9 in init-workspace-flow
- Input: state.mode, state.plugin_active, IDEs
- Output: LSP/codegraph configs, CONTEXT.md additions
- Prerequisite: Phase 1-5 complete, state.mode set </workflow_context>
<phase_steps>
- Check state
- Learn
- Suggest LSPs, codegraphs
- Update state with phase status </phase_steps>
<check_mode step="6.1">
- Read
agents/init-workspace-flow-state.md - Check if LSPs or GitNexus or Graphify is already installed
- If upgrade mode: check whether user wants to change anything => basically confirm and allow to switch
</check_mode>
<learn_codegraph step="6.2">
- Default —
CODEMAP.md: built-in, no install, no third party, no cost, works on <300 source code files projects very easily. - Graphify (
https://github.com/safishamsi/graphify): MIT-licensed, free. - GitNexus (
https://github.com/abhigyanpatwari/GitNexus): free for non-commercial or personal use, PAID for commercial or business use. Likely this will be a blocker. - LSP: locally running lightweight Language Server Protocol servers for different programming languages
- Exclusive selection: Graphify XOR GitNexus XOR LSPs. Multiple LSPs allowed. Default is perfect with LSPs.
- Each IDE/coding agent requires different setup, there are SPECIFIC commands/params.
- CRITICAL: our integration requires tools/LSPs to be on the PATH or available by just using tool name without paths.
- IMPORTANT: user must do initialization for any extra, like "graphify .", etc.
</learn_codegraph>
<codegraph_known_links notice="DO NOT READ IN ADVANCE">
- GitNexus: https://raw.githubusercontent.com/abhigyanpatwari/GitNexus/refs/heads/main/README.md
- GitNexus Licensing: https://github.com/abhigyanpatwari/GitNexus?tab=readme-ov-file#enterprise
- Graphify: https://raw.githubusercontent.com/safishamsi/graphify/HEAD/README.md
- Claude Code:
- User must install the language server binary separately.
- User installs them from the official marketplace: search for lsp keyword using the
/plugincommand Discover tab. - List of available: typescript-lsp, pyright-lsp, csharp-lsp, gopls-lsp, rust-analyzer-lsp, jdtls-lsp (Java Eclipse JDT.LS), php-lsp, clangd-lsp, kotlin-lsp, swift-lsp, lua-lsp, ruby-lsp, liquid-lsp (Shopify Liquid), clojure-lsp
- Claude code intentionally support only those LSPs as those are stable, others may not
- More available (with actual instructions for binaries): https://raw.githubusercontent.com/Piebald-AI/claude-code-lsps/refs/heads/main/README.md
- Codex does NOT support LSPs
- Copilot CLI LSPs: https://docs.github.com/en/copilot/how-tos/copilot-cli/set-up-copilot-cli/add-lsp-servers (only CLI and CLI mode, not regular VS Code and Jetbrains IDE copilots)
</codegraph_known_links>
<suggest_codegraph step="6.3">
- Read docs/TECHSTACK.md if not in context, to be grounded
- Always show selector with recommend Default first, then optional LSPs, then Graphify, then GitNexus (show ALL even if something is installed)
- Warn the user about licenses and IP: "third-party tool will have access to IP", "review license and policy with your manager"
- Explain you will NOT install yourself, but if installed IT IS supported by Rosetta
- Explain that they need to add "MUST USE SKILL CODEMAP AND FOR CODE NAVIGATION" to docs/CONTEXT.md at the end so that AI knows to use it (note skill already handles)
- Upon selection - do NOT install, read, search, understand paying attention to IDE/coding agent and languages
- Only then provide very brief, concise, specific, and short instructions for user to install plus exact deep links, make this guidance to visually standout from overall log of all activities
- You can guide user too AFTER entire workflow is completed, but do NOT install yourself, you can modify CONTEXT.md too
</suggest_codegraph>
<update_state step="6.4">
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today First seen · 91 lines · 18 tokens per session scan A a19ac3ab4545
init-workspace-flow-codegraph is a command published in the GitHub repository griddynamics/rosetta (342 stars, last pushed today), licensed Apache-2.0. It adds 18 tokens to every session and 1,163 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other commands, from other repositories
resolve-conflicts
Resolve merge conflicts on a GitHub PR by merging the base branch into the PR branch in an isolated git worktree, with JSON/markdown-aware conflict resolution.
ship
Commit all changes and push to remote in one operation (combines /commit and /push).
react-router-client-loader
Do this in a clientLoader and use loaderData to render the component. DO NOT create mock data, new interfaces, or mock data loader functions. Instead, assume loaderData has all of the data you need to render the component.
OPSX: Apply
Implement tasks from an OpenSpec change (Experimental).
deploy-check
Pre-flight verification before pushing to main. Reproduces the CI gates locally so the push doesn't fail in production.
hook-install
Install a git hook for automatic commit sync on push. When installed, every git push automatically syncs pushed commits to Avos Memory, enabling real-time team synchronization.