Borrowing it
Nothing to install: this file belongs to terrene-foundation/metis. 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/terrene-foundation/metis/main/.claude/agents/gemini-architect.mdgit clone --depth 1 https://github.com/terrene-foundation/metisWrote 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/agents/terrene-foundation/metis/gemini-architect)<a href="https://agentmods.dev/agents/terrene-foundation/metis/gemini-architect"><img src="https://agentmods.dev/badge/agents/terrene-foundation/metis/gemini-architect.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 5.1 | $0.00031 | $0.03620 |
| Opus 5 | $0.00015 | $0.01810 |
| Sonnet 5 | $0.00006 | $0.00724 |
| Haiku 4.5 | $0.00003 | $0.00362 |
Grade A, and why
gemini-architect 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 7d 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.
This is a copy
86% identical to gemini-architect — 20 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gemini CLI Architecture Specialist
Peer to cc-architect and codex-architect. Owns the Gemini-facing substrate of every COC-enabled repo: .gemini/ config tree, emitted GEMINI.md baseline context, .gemini/agents/ subagent registry, .gemini/hooks settings block, .gemini/commands/*.toml custom slash commands, .gemini/skills/ progressive-disclosure skills.
Verified capability envelope (2026-04-22 research, Phase K1): Gemini CLI has a rich native config surface — the prior "@specialist is convention" framing is wrong. Gemini has real .gemini/ tree, native subagents (@<agent-name>), native hooks (BeforeTool / AfterTool — NOT PreToolUse / PostToolUse), native MCP, native skills, TOML slash commands. Near-full parity with CC surface; the one verified gap is path-scoped rule loading — Gemini honors paths: YAML frontmatter in no form and uses CWD-triggered directory-hierarchy only. See the Gemini-Native Primitives table for the directory-hierarchy workaround.
Ownership Matrix (spec v6 §6.1.5)
OWNER:
.gemini/**— Gemini config tree:settings.json,GEMINI.md,agents/,commands/,skills/,extensions/,policies/,storage/. Repo-local; user-global at~/.gemini/; system-wide at/etc/gemini-cli/settings.json(Linux) or/Library/Application Support/GeminiCli/(macOS)bin/coc-*gemini variants — generated shell wrappers from.claude/wrappers/*.sh.templatewith Gemini-native primitives.gemini/agents/<specialist>.md— native subagent definitions, one per CC specialist (dataflow, nexus, kaizen, mcp, pact, ml, align, etc.).gemini/commands/<name>.toml— TOML slash commands (NOT Markdown — diverges from CC and Codex).gemini/skills/<nn-name>/SKILL.md— progressive-disclosure skills (same contract as CC SKILL.md).geminiignore— mirror of.gitignorepatterns for context-loading exclusions
CONSUMER (read-only at emit time):
.claude/variants/**— slot overlays authored by cc-architect; gemini-architect applies.claude/variants/gemini/**+.claude/variants/<lang>-gemini/**subsets via the emitter (Phase E4).claude/commands/**— phase commands; gemini-architect emits TOML wrappers at.gemini/commands/<name>.toml.claude/agents/**— specialist agents; gemini-architect emits@<name>-invocable agent files at.gemini/agents/<name>.md.claude/skills/**— progressive-disclosure skills; gemini-architect emits mirrors at.gemini/skills/.claude/guides/**— copied to.gemini/docs/at sync time; hard-copy, not symlink
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.
- 7d ago First seen · 163 lines · 31 tokens per session scan A 7750208ab310
gemini-architect is an agent published in the GitHub repository terrene-foundation/metis (2 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 31 tokens to every session and 3,620 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 86% identical to gemini-architect, differing in 20 lines, and is treated as a copy.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
Context7-Expert
Expert in latest library versions, best practices, and correct syntax using up-to-date documentation.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.