1---
2name: researching-with-rumilo
3description: Dispatches AI research subagents via rumilo to search the web or explore git repositories. Use when the user asks to research a topic, look up library usage, explore a codebase, read documentation from the web, or when you need information from an external repository or website to complete a task.
4license: GPL-3.0-or-later
5metadata:
6 author: Amolith <amolith@secluded.site>
7---
8
9`rumilo` spawns disposable AI subagents for two kinds of research: **web search** and **repo exploration**. Each mode runs in isolation and returns a synthesised answer on stdout.
10
11## Important: timeouts
12
13Rúmilo is an LLM-based subagent that may take multiple turns to complete its research. _Never_ impose a timeout.
14
15## Modes
16
17### Web mode
18
19Search the internet or ask questions about a specific page.
20
21```bash
22# Open-ended search (no URL)
23rumilo web "what is the current stable version of Zig"
24
25# Question about a specific page (URL pre-fetched into context)
26rumilo web "how does the slog package handle log levels" -u "https://pkg.go.dev/log/slog"
27```
28
29- Without `-u`, the agent searches the web on its own.
30- With `-u`, the URL is fetched first and injected into the agent's context alongside the query, so it can answer questions _about_ that page.
31
32### Repo mode
33
34Clone a git repository into a temp directory, spawn an agent inside it, and let it explore the source to answer questions.
35
36```bash
37# Public repo
38rumilo repo -u "https://github.com/owner/repo" "How does the routing system work?"
39
40# Private repo (works with any clonable URI — SSH, HTTPS with auth, etc.)
41rumilo repo -u "git@git.sr.ht:~user/repo" "What does the Config struct look like?"
42```
43
44The `-u` flag is **required** in repo mode — it needs something to clone.
45
46## Flags
47
48| Flag | Modes | Effect |
49| -------------------------- | ----- | ----------------------------------------------------------------- |
50| `-u <url>` | both | Seed URL (web) or clone URI (repo) |
51| `--model <provider:model>` | both | Override the configured default model |
52| `--ref <ref>` | repo | Checkout a branch, tag, or commit after cloning (requires --full) |
53| `--full` | repo | Full clone instead of shallow (depth 1) |
54| `--no-cleanup` | both | Keep the temp workspace after exit |
55| `--verbose` | both | Print tool calls and results to stderr |
56
57## Crafting queries
58
59Rumilo works best with focused queries. Think about **scope** and **coupling** when deciding what to ask and how many questions to include.
60
61- **Tightly coupled, narrow scope** — multiple simple questions in one call is fine when they're about the same small surface area (a single file, one struct, one API endpoint).
62- **Broad scope or loose coupling** — one question per invocation. If answering requires exploring many files or pages, keep it to a single clear question so the agent can focus.
63- **Write natural questions, not keyword lists.** Rumilo is an LLM, not a search index.
64- **Sequence calls when later questions depend on earlier answers.** Get the overview first, then drill in. Rumilo is amnesic — each invocation starts fresh. If a later query depends on an earlier answer (repo URL, version number, file paths), you must include that context explicitly in the prompt.
65- **Run independent queries in parallel** (e.g. via separate tmux windows or backgrounded commands), but keep dependent queries sequential.
66
67### Bad
68
69Overloaded — five loosely-coupled questions spanning an entire diff, each requiring different context:
70
71```bash
72rumilo web "In this diff, find: 1. What files are modified and created \
732. What config options each channel exposes 3. How messages are formatted \
744. The architecture for adding a new channel 5. What UI components exist" \
75 -u "https://example.com/commit/abc123.diff"
76```
77
78### Good
79
80Sequenced calls, each with a clear scope. Later calls include non-obvious context from earlier results — file paths, symbol names, patterns — so rumilo can go straight to the answer instead of searching.
81
82```bash
83# 1. Lay of the land — broad overview first
84rumilo web "What files were modified or created, and what's the broad shape of the change?" \
85 -u "https://example.com/commit/abc123.diff"
86
87# 2a and 2b can run in parallel — neither depends on the other,
88# but both benefit from context learned in step 1.
89
90# 2a. Architecture deep-dive, pointing to the key file discovered in step 1
91rumilo web "How is the notification system in pkg/notify/ architected? \
92 Each channel seems to implement a Sender interface — what's the process \
93 for adding a new one?" \
94 -u "https://example.com/commit/abc123.diff"
95
96# 2b. Config drill-down, referencing the settings struct found in step 1
97rumilo web "What configuration options does each channel expose in \
98 ChannelSettings? Can message content be customised?" \
99 -u "https://example.com/commit/abc123.diff"
100
101# 3. UI specifics, narrowed by what steps 1–2 revealed about the component tree
102rumilo web "The channel config UI lives in web/src/components/settings/channels/. \
103 Do the per-channel forms follow a shared pattern or component?" \
104 -u "https://example.com/commit/abc123.diff"
105```
106
107Don't include context that's obvious or redundant — telling rumilo to "search for notif" when asking about notifications wastes tokens. Focus on non-obvious anchors: exact file paths, struct names, interface signatures.
108
109Tightly coupled questions in a single call are fine when the scope is small:
110
111```bash
112rumilo repo -u "https://github.com/owner/repo" \
113 "What fields does the Config struct have and which are optional?"
114```
115
116## Choosing a mode
117
118- Need to find or learn about a library? **Web** to locate it, then **repo** to explore its source.
119- Need to understand how a dependency works in detail? **Repo** with the dependency's clone URL.
120- Need current docs, release notes, or general knowledge? **Web**, optionally with the URL of a relevant page.