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 skills add QuipNetwork/qubitrefill --skill qupickgit clone --depth 1 https://github.com/QuipNetwork/qubitrefillWrote 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/skills/quipnetwork/qubitrefill/qupick)<a href="https://agentmods.dev/skills/quipnetwork/qubitrefill/qupick"><img src="https://agentmods.dev/badge/skills/quipnetwork/qubitrefill/qupick/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/skills/quipnetwork/qubitrefill/qupick"><img src="https://agentmods.dev/badge/skills/quipnetwork/qubitrefill/qupick.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 | $0.00023 | $0.07162 |
| Opus 5.5 | $0.00009 | $0.02865 |
| Sonnet 5.5 | $0.00005 | $0.01432 |
| Haiku 4.5 | $0.00002 | $0.00716 |
Grade B, and why
qupick scanned grade B with 2 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 2mo 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.
Reads MCP configurationmediumAgent snooping
mcp.json carries server URLs and auth tokens; reading it lets a mod discover and abuse other integrations.
- Use a dedicated low-balance account. `config.json` (real email) and the API key in `.mcp.json` (the agent's secret key) are local-only — both files are gitignored, so the key never goes in git. Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`mcp__qupick__*` tools — **no `curl`**. The six tools are allowlisted, so none of them prompt: How it starts
The opening of the file, as written. The whole thing — 465 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pay with most suitable crypto asset in your portfolio
Identify the most suitable crypto in the portfolio (lowest annualised expected return) using a quantum unconstrained binary optimization, settle a Bitrefill product against the cheapest available funding source, then retune the portfolio only if the chosen crypto was actually sold.
Delegates all purchase mechanics to the Bitrefill MCP (mcp__bitrefill__* tools) — connect it separately (claude mcp add --transport http bitrefill https://api.bitrefill.com/mcp, or install the upstream bitrefill plugin which bundles it; https://github.com/bitrefill/agents), then invoke it for product search, pricing, buying, and payment polling. This skill adds portfolio seeding, selection logic, and an account-aware funding waterfall on top.
The flow is designed to stop for the user in exactly one place — the purchase approval (step 6). Defaults, a config file, and a permission allowlist remove the other interruptions.
Calling conventions (MCP tools)
The backend is the qupick MCP server — the portfolio backend's /mcp transport. Its URL is
whatever .mcp.json points the qupick server at: http://127.0.0.1:8000/mcp for a local stack,
or a deployed HTTPS endpoint such as https://qupick.quip.network/mcp. Drive it with
mcp__qupick__* tools — no curl. The six tools are allowlisted, so none of them prompt:
| Tool | Does |
|---|---|
mcp__qupick__ping_backend |
liveness probe (public) |
mcp__qupick__register_agent |
create an agent; the API key is emailed (public) |
mcp__qupick__get_agent |
fetch the authenticated agent's config (basket, sliders) |
mcp__qupick__optimize |
optimise / retune the authenticated agent |
mcp__qupick__get_market |
live holdings + μ for the authenticated agent |
mcp__qupick__get_leaderboard |
scoreboard (public) |
Auth. Every per-agent tool (get_agent, optimize, get_market) carries the agent's API
key as Authorization: Bearer <key>, configured once in .mcp.json by pasting the key literally
into the header ("Authorization": "Bearer <key>"). Claude Code does not expand ${VAR} in
.mcp.json headers, so an environment variable won't reach the server — use the literal key from
registration (see step 2). The public tools work without it.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2mo ago First seen · 465 lines · 23 tokens per session scan B 0085373a5529
qupick is a skill published in the GitHub repository QuipNetwork/qubitrefill (5 stars, last pushed 18d ago), licensed Apache-2.0. It adds 23 tokens to every session and 7,162 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it B with 2 findings (reads mcp configuration, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
qupick compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| qupickQuipNetwork/qubitrefill | Skill | 5 | 18d ago | Apache-2.0 | B | 23 | Claude Code, Codex |
| agentic-streetLeoYeAI/openclaw-master-skills | Skill | 2.2k | 6d ago | MIT | A | 91 | Claude Code, Codex |
| token-pickaeonfun/aeon | Skill | 788 | yesterday | MIT | A | 25 | Claude Code, Codex |
| bridgealsk1992/CloddsBot | Skill | 3.1k | 1mo ago | MIT | A | 13 | Claude Code, Codex |
| hyperliquidalsk1992/CloddsBot | Skill | 3.1k | 1mo ago | MIT | A | 19 | Claude Code, Codex |
| marginfialsk1992/CloddsBot | Skill | 3.1k | 1mo ago | MIT | A | 11 | Claude Code, Codex |
| airdrop-hunting-guidenirholas/three.ws | Skill | 269 | yesterday | Apache-2.0 | A | 54 | Claude Code, Codex |