Write code with Niro
While code is being written, the coding agent asks the graph instead of searching through files. One question to the graph can replace many text searches and file reads. For example: "who calls this method, three levels up?"
Who and how
- The coding agent reaches Niro over MCP (the Model Context Protocol, the standard way coding agents use outside tools). Niro sets this up for 14 agents. See Connect your coding agent to Niro.
- The developer asks from the terminal with
niro ask. Run it with no tool name to list the tools.
What agents are taught to do
The skill Niro installs tells the agent to do four things:
- Explore first. Use
code_grep,get_source_code,resolve_name,list_types,get_class_structure,find_api_endpointsandfind_handlerbefore reading files by hand. - Check both directions before changing anything. Use
get_blast_radius,get_reverse_call_chainandget_impact_analysis. - Follow flows across services. Use the HTTP, event and data tools.
- Say when Niro does not apply. For example: the folder is not indexed, or the graph is behind the files.
Claude Code gets more help than the other agents. Niro adds hooks to Claude Code: small commands that Claude Code runs at fixed moments. One of them shows an "impact check" before Claude Code edits a file. It is advisory only: it shows the information and never blocks the edit. To turn it off, run touch ~/.niro/impact-gate-off. The other agents get instruction text instead of hooks.
When the graph is behind your edits
The graph is rebuilt in the background about 3 seconds after a folder stops changing. If an answer is about a file that changed after the last build, the answer says so, and lists the files to read directly.
Playbooks for this stage
- Change code safely: callers before an edit, overloads, static imports, common names grep cannot narrow.
- Add a feature: the interface to implement, its methods, existing implementations to copy.
- Ask for less:
answer=count,locations,names,existsandmatch, so an answer carries only what the next step needs.