Borrowing it
Nothing to install: this file belongs to llama-farm/llamafarm. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/llama-farm/llamafarm/main/.agents/commands/research.mdgit clone --depth 1 https://github.com/llama-farm/llamafarmWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/llama-farm/llamafarm/research)<a href="https://agentmods.dev/commands/llama-farm/llamafarm/research"><img src="https://agentmods.dev/badge/commands/llama-farm/llamafarm/research/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/llama-farm/llamafarm/research"><img src="https://agentmods.dev/badge/commands/llama-farm/llamafarm/research.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00000 | $0.01488 |
| Opus 5 | $0.00000 | $0.00744 |
| Sonnet 5 | $0.00000 | $0.00298 |
| Haiku 4.5 | $0.00000 | $0.00149 |
Grade A, and why
research 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 9d ago.
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 — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codebase Research Playbook
A fast, reliable way for agents to understand this monorepo before implementing an issue. Follow this checklist to map scope, find the right code, and de‑risk changes.
Objectives
- Understand the issue: clarify scope, inputs/outputs, and success criteria.
- Locate the code path: find entry points, flows, configs, and tests.
- De‑risk: identify edge cases, cross‑project impacts, and verification steps.
Quickstart Checklist (15–30 min)
- Read the issue: capture user story, acceptance criteria, affected area(s), and constraints.
- Identify project(s): server, cli, rag, models, docs. If unsure, start with Nx graph.
- Skim module entry points:
server/main.pyandserver/api/**cli/main.goandcli/cmd/**rag/cli.pyandrag/**docs/website/**
- Find existing tests near the suspected code.
- Collect commands to run (build/test/dev) so you can validate later.
Workspace-first view (Nx preferred)
If Nx tools are available, prefer them to map dependencies and tasks.
- Graph: visualize projects and edges.
- Example: run a project graph or open the visualization tool to see producers/consumers.
- Show project details: inspect target configurations (build/test/lint/dev).
- Docs: consult relevant Nx docs for generators/tasks if needed.
If Nx tools are not available, manually inspect:
nx.json,project.jsonfiles at repo and project rootspackage.jsonscripts (if present in subprojects)server/project.json,server/pyproject.tomlcli/go.mod,cli/project.jsonrag/pyproject.toml,rag/default_strategies.yaml
How to search effectively
- Start semantic, then go exact:
- Use semantic search for “how/where” questions (endpoints, flows, responsibilities).
- Use exact search (ripgrep/grep) for symbols, function names, routes, or constants.
- Good semantic queries:
- “Where is user authentication handled?”
- “How are CLI commands registered?”
- “Where is RAG document ingestion triggered?”
- Good exact searches (ripgrep examples):
rg --hidden --glob '!**/dist/**' --glob '!**/node_modules/**' "uvicorn|FastAPI" server | cat
rg --hidden "cobra\.Command" cli | cat
rg --hidden "pytest|conftest\.py" server | cat
rg --hidden "project\.json" -n | cat
- Trace the flow: when you find a hit, read nearby files; follow imports, routes, and callers.
- Expand outward until you can explain inputs, transformations, side effects, and outputs.
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.
- 9d ago First seen · 134 lines · 0 tokens per session scan A acb8b031775c
research is a command published in the GitHub repository llama-farm/llamafarm (838 stars, last pushed 3mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,488 tokens. 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 2026-08-30.
Other commands, from other repositories
model-config
Configure AI provider models for Claude Octopus workflows.
model-check
Reassess the current session and recommend the right model and effort level.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.