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 instructions/rmedranollamas/research-cli/agents-mdgit clone --depth 1 https://github.com/rmedranollamas/research-cliWrote 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/instructions/rmedranollamas/research-cli/agents-md)<a href="https://agentmods.dev/instructions/rmedranollamas/research-cli/agents-md"><img src="https://agentmods.dev/badge/instructions/rmedranollamas/research-cli/agents-md.svg" alt="Measured on agentmods" 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 | $0.00549 | $0.00549 |
| Opus 5 | $0.00275 | $0.00275 |
| Sonnet 5 | $0.00110 | $0.00110 |
| Haiku 4.5 | $0.00055 | $0.00055 |
Grade A, and why
research-cli AGENTS.md 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 4d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Notes
research-cli is a Go CLI for Gemini Deep Research and multimodal interactions.
Architecture
- Deep Research uses
deep-research-preview-04-2026through the Gemini v1alpha Interactions API. - Max Deep Research can be selected with
deep-research-max-preview-04-2026. - Fast search uses
gemini-3-flash-preview. - Image generation uses
gemini-3-pro-image-preview. - Long-running interactions stream over SSE, then fall back to polling if the stream ends before a report is available.
- Local task history is stored in SQLite at
~/.research-cli/history.dbby default.
Gemini CLI Extension
- Manifest:
gemini-extension.json - Slash command:
commands/research.toml - Skill:
skills/gemini-research/SKILL.md - Release archives place the platform binary at
skills/gemini-research/scripts/research.
Security Notes
RESEARCH_GEMINI_API_KEYis passed only to Gemini API requests.- Custom remote API base URLs must use HTTPS. HTTP is accepted only for real loopback hosts.
- Workspace path validation rejects parent traversal and symlink output overwrites.
- Untrusted streamed model text is stripped of terminal control sequences before printing.
- The SQLite history database is created with
0600permissions.
Development
go test ./...
go vet ./...
go build ./...
The workspace has used Go 1.26.3. Keep the go directive in go.mod aligned
with the intended release toolchain; the release workflow uses
actions/setup-go with go-version-file: go.mod.
Coverage:
go test ./... -coverprofile=coverage.out
go tool cover -func=coverage.out
Release
Pushing a v* tag runs .github/workflows/release.yml, which builds Go binaries for Linux and macOS on amd64 and arm64, packages Gemini extension archives, and publishes checksums.
After a release, verify at least one archive by downloading it with
gh release download, comparing its SHA-256 against checksums.txt, extracting
skills/gemini-research/scripts/research, and checking research --version.
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.
- 4d ago First seen · 62 lines · 549 tokens per session scan A 6d708c33202c
research-cli AGENTS.md is an instructions file published in the GitHub repository rmedranollamas/research-cli (20 stars, last pushed 4d ago), licensed Apache-2.0. It adds 549 tokens to every session, about $0.0027 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 2026-08-30.
Other instructions, from other repositories
gryph AGENTS.md
Instructions for safedep/gryph, covering gryph, build, architecture, key entry points and notes.
coral CLAUDE.md
Instructions for cdknorow/coral, covering claude.md - coral go, mission, testing, go unit tests and legacy parity tools (historical reference).
gemini-cli-extension GEMINI.md
Instructions for pinecone-io/gemini-cli-extension, covering pinecone extension for gemini cli, available agent skills, key concepts & setup and available mcp tools.
gemini-faf-mcp GEMINI.md
Instructions for Wolfe-Jam/gemini-faf-mcp, covering gemini.md — gemini-faf-mcp, test & verify, where things live, stack and before changing things.
everything-gemini-code GEMINI.md
Gemini CLI instructions for Jamkris/everything-gemini-code, covering project instructions, prompt defense baseline and guidelines.
google-ads-gemini-extension GEMINI.md
Instructions for itallstartedwithaidea/google-ads-gemini-extension, covering google ads agent extension, mcp tools — read & analysis, mcp tools — write (gated, confirmation required), mcp tools — auth (method 2) and slash commands.