Is fugu-second-opinion safe to install?

SaferSkills independently audited fugu-second-opinion (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

CategoryWeightCategory scoreContribution

Securityprompt, exec, net, exfil, eval

35%

100

35.0 pts

Supply chainhash, typosquat, maintainer, lockfile

20%

100

20.0 pts

Maintenancestaleness, pinning, CI

15%

100

15.0 pts

TransparencySKILL.md, perms, README

15%

100

15.0 pts

Communityinstalls, verify, response

15%

100

15.0 pts

Findings & checks · 0 flagged

Securityscore 100 · 0 findings

✓—All security checks passedNo findings in this category for the latest scan.pass

Supply chainscore 100 · 0 findings

✓—All supply chain checks passedNo findings in this category for the latest scan.pass

Maintenancescore 100 · 0 findings

✓—All maintenance checks passedNo findings in this category for the latest scan.pass

Transparencyscore 100 · 0 findings

✓—All transparency checks passedNo findings in this category for the latest scan.pass

Communityscore 100 · 0 findings

✓—All community checks passedNo findings in this category for the latest scan.pass

The primary manifest — the file an agent reads to learn what this artifact does.

skill/SKILL.md · 1 file

skill/SKILL.md1.9 KB · Markdown

name: fugu-second-opinion

description: >-

Get a multi-model "second opinion" from Sakana Fugu (an orchestration model

that routes across GPT/Claude/Gemini) on hard, self-contained problems, or to

cross-check a risky decision. Backed by the fugu-mcp MCP server's `ask_fugu`

and `fugu_second_opinion` tools. Use for tough isolated reasoning — NOT for

routine, interactive, or iterative repo work (that's slower and pricier here).

metadata:

homepage: https://github.com/jamesaoyama/fugu-mcp

requires:

binaries: [node]

env: [SAKANA_API_KEY]

mcp:

name: fugu

install: "openclaw mcp add fugu --command node --arg /path/to/fugu-mcp/dist/index.js --env SAKANA_API_KEY=YOUR_KEY"

Fugu second opinion

fugu-mcp exposes Sakana Fugu — a multi-model orchestration model — as two MCP tools your agent can call on demand.

When to use it

A hard, self-contained reasoning/coding problem where a multi-model

ensemble may beat one model.

Cross-checking a risky or high-stakes decision (a second, independent take).

When NOT to use it

Routine, interactive, or iterative work in your repo — that's what your main

agent loop is for. Fugu adds latency (30–120s) and cost (~$30/1M output, plus billed orchestration tokens).

Tools

ask_fugu — { prompt, system_prompt?, model?, max_tokens? }. General hard

query / second opinion.

fugu_second_opinion — { content, question, model? }. Hand it code/an

answer + a question; get back a rigorous critique.

Setup

Build: git clone https://github.com/jamesaoyama/fugu-mcp && cd fugu-mcp && npm install && npm run build
Get a key from the Sakana console; export SAKANA_API_KEY.
Register with OpenClaw:

openclaw mcp add fugu --command node --arg "$(pwd)/dist/index.js" --env SAKANA_API_KEY=YOUR_KEY

Confirm: openclaw mcp list, then openclaw mcp reload.

See the repo README for Claude Code setup and full config options.

fugu-second-opinion— agent skill

Is fugu-second-opinion safe to install?

Fugu second opinion

When to use it

When NOT to use it

Tools

Setup

Audit the pieces. Scan the whole. Decide.