What is fusion-status?

A command that checks whether the Fusion setup is installed, loaded, and enforcing its rules. Fusion is an opencode agent-team configuration.

It is for checking the live session and configuration files when diagnosing Fusion setup problems.

It shows which parts of the setup pass or fail, with evidence, without changing anything.

fusion-status — Command for coding agents by mihneaptu: A command that checks whether the Fusion setup is…

Install

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.

agentmods

npx agentmods add commands/mihneaptu/opencode-fusion/fusion-status

Clone the repo

git clone --depth 1 https://github.com/mihneaptu/opencode-fusion

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/mihneaptu/opencode-fusion/fusion-status.svg)](https://agentmods.dev/commands/mihneaptu/opencode-fusion/fusion-status)

Your own site

<a href="https://agentmods.dev/commands/mihneaptu/opencode-fusion/fusion-status"><img src="https://agentmods.dev/badge/commands/mihneaptu/opencode-fusion/fusion-status.svg" alt="Measured on agentmods" height="20"></a>

Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 810 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00012	$0.00810
Opus 5	$0.00006	$0.00405
Sonnet 5	$0.00002	$0.00162
Haiku 4.5	$0.00001	$0.00081

Measured 5d ago against content hash 0d329e79e23c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

fusion-status 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 5d 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.

.opencode/skills/fusion-setup/commands/fusion-status.md · 21 lines

What it actually says

Run a Fusion health check and report the results. Check four layers, report each enabled layer as PASS or FAIL with one line of evidence, and do NOT fix anything - this command only reports.

Throughout, CONFIG DIR means opencode's global config directory: $XDG_CONFIG_HOME/opencode/ when XDG_CONFIG_HOME is set and non-empty, otherwise ~/.config/opencode/ (Windows: %USERPROFILE%\.config\opencode\). The paths below are written with the common default; resolve them against the real value first, or a correct install under a custom XDG_CONFIG_HOME will look missing and be reported as a false FAIL.

Live enforcement (this running session). Inspect your own available tools - do not call anything, just check what exists in your toolset. Fusion denies edit, grep, glob, and list for this agent, and opencode removes denied tools from the tool schema entirely. If any of those four tools is available to you right now, the Fusion config is NOT loaded in this session - report FAIL and that a full restart of opencode is the fix. If all four are absent and task is available, report PASS.
Config on disk. Read ~/.config/opencode/opencode.json (Windows: %USERPROFILE%\.config\opencode\opencode.json). FAIL if the file is missing or not valid JSON. Otherwise list each agent.<role>.model assignment found, and FAIL if build or sidekick has no model (explore should normally have one too; research/design/reviewer/vision are optional). Then read ~/.config/opencode/.fusion-install.json and report bundleVersion and installedAt as the bundle applied most recently - a reapply that selected only some roles can leave other managed files from an earlier bundle, so report it as the last apply rather than as a guarantee about every file. Report unknown (pre-1.1.0) when the manifest exists but has no bundleVersion, and not installed by the installer when the manifest is absent - a hand-copied setup has no manifest, so that alone is not a FAIL.
Agent files. For every role that has a model in the config - plus plan, minus explore (explore is built-in and needs no file) - read ~/.config/opencode/agent/<role>.md and confirm the file exists. In build.md, confirm the frontmatter contains edit: deny. A missing file or a missing denial is FAIL, with the exact path.
Optional Claude Code bridge. Treat this layer as disabled and report SKIP only when both ~/.config/opencode/plugins/fusion-claude.js and the top-level permission.fusion_claude_* rule are absent. Otherwise confirm the plugin file exists, the top-level rule is deny, both build.md and plan.md allow fusion_claude_status and fusion_claude_review, and both tools are present in this build agent's live toolset. If those checks pass, call fusion_claude_status and report its sanitized result. Never run claude auth status directly because its output may contain account identity data.

Diagnosis rules: layer 1 FAIL with layers 2-3 PASS means the config changed after startup - a restart fixes it. A role with a model in config but no agent file (except explore) runs without its Fusion permissions and prompt - point to the fusion-setup skill to install it.

End with a one-line verdict: Fusion: healthy or Fusion: <n> issue(s) followed by the specific fix for each issue.

Changes

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.

5d ago First seen · 21 lines · 12 tokens per session scan A 0d329e79e23c

Subscribe to this mod's changes

fusion-status is a command published in the GitHub repository mihneaptu/opencode-fusion (216 stars, last pushed 11d ago), licensed MIT. It adds 12 tokens to every session and 810 once invoked, about $0.0001 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.

Multi-Agent ai-agents ai-coding model-routing opencode opencode-config

fusion-status

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

What this file has done since we first saw it

Other commands, from other repositories

models-preset

session-info

session-id

vscode

swarm

distill