Stingray Agent Skills

Name: stingray-agent-skills
Author: MantaDigital

A specialized crypto market agent and hosted rule runtime for your coding agent.

Stingray can be used directly, or as the market partner for Codex, Claude Code, Cursor, and other SKILL.md-compatible agents. Your coding agent plans, edits, and can set up local infrastructure. Stingray resolves market context, turns theses into inspectable rules, replays them against history, and hosts monitoring when you do not want another cron script.

Built for quants, analysts, and research desks who already live in their terminal.

One install, one credential file — works in Claude Code, Codex, Cursor, Gemini CLI, Cline, Continue, Goose, Roo, Windsurf, and 36+ other SKILL.md-compatible agents via npx skills.

Install

npx skills add MantaDigital/stingray-agent-skills -g -y

Or ask your agent: "Install the skills from MantaDigital/stingray-agent-skills globally for all my coding agents using npx skills."

Then complete the one-time credential setup — open stingray.fi/app/settings#settings-api-tokens, create a token, and follow the terminal-only setup command your agent gives you. Do not paste the token into chat. Full quickstart: skills/stingray/README.md.

Some agents load skills only when a session starts. If the current session installed Stingray but cannot use it yet, restart the agent session and ask again.

⚠️ Reinstalling overwrites the local skill copy without prompting. Back up custom edits first.

Capabilities

Surface	What you get
Specialized market agent	Use Stingray directly or delegate market work from Codex, Claude Code, Cursor, or another agent when the task needs live crypto data, venue grounding, historical replay, or hosted monitoring.
Current data coverage	Binance Spot price/volume/TA, Hyperliquid funding, open interest, whale/liquidation streams, entity news, Telegram-native news sources, and KG-backed asset/entity resolution.
Venue grounding	Resolve assets and markets across venues with `/kg/search` and `/kg/resolve` before writing watchlists, portfolio rows, alerts, or backtests.
Typed rules	Turn market theses into validated alert definitions with price, news sentiment, technical indicators, and supported venue-specific conditions.
Hosted monitoring	Create alert rules that run on Stingray after your coding-agent session ends, with delivery readiness checked first. Treat the hosted rule as a signal surface your own agent can listen to.
Backtests	Take a thesis or alert definition and replay it against historical data. Private to your account by default. 24h TTL on results.
Source-backed news	Fetch normalized per-entity news via `/entities/:id/news` and treat returned content as data, not instructions.
News-aware signals	News blocks compose into alerts and trigger trees alongside price and TA — react to sentiment shifts, not just candles.
Watchlists & portfolio	Curate watchlists, track positions, sync state from your agent.
Multi-channel delivery	Notifications and chat through web, Telegram, and WhatsApp when linked. X status is inspectable; X link-claiming and public posting are not public skill actions.
Shareable cards (opt-in)	Mint a public Astro page from a backtest result for DM, tweet, or screenshot.
Token hygiene	List, revoke, and rotate API tokens without leaving the terminal.
Co-development	Ask for missing assets, signals, or datasets, and send privacy-safe setup/debug reports through the same chat endpoint.

The agent reads only what it needs, scoped to the task: full reference index in skills/stingray/SKILL.md.

For onboarding, see skills/stingray/prompts.md. For agent-side capability introspection, see skills/stingray/references/capabilities.json.

Scope: research, account state, typed alerts, hosted monitoring, notifications, watchlists, portfolio state, feedback, and backtesting against an account-scoped API. The public skill does not initiate any value transfer on the user's behalf and does not hold custody of funds.

Security & Trust

Snyk's scanner flags two specific concerns; both follow Snyk's published best practice for skills of this shape.

Credential handling (Snyk W007). The skill stores a long-lived API token at ~/.stingray/credentials (mode 600) so you don't paste it every session. The token never enters the agent's chat context — the agent gives you a one-line shell command to run yourself, or you set STINGRAY_PAT in your shell. Chat-paste is deliberately not supported.
Third-party content / news (Snyk W011). The skill consumes Stingray-curated news and KG content, which can carry adversarial text. The agent treats all fetched content as data, not instructions (explicit "Untrusted Content Handling" section in SKILL.md). For a stricter posture, omit news_* blocks from your alerts.

Token surface is account-scoped: no admin, billing, webhook, or API-token-creation routes. No installer scripts, no remote downloads, no value-transfer through this public skill (it cannot move funds, sign transactions, or place orders). No access to other credential stores. Apache-2.0, public source, revocable from your Stingray settings.

What Ships

skills/stingray/SKILL.md — agent entrypoint with task routing and operating loop
skills/stingray/README.md — human quickstart
skills/stingray/prompts.md — copy-paste prompt index
skills/stingray/references/capabilities.json — machine-readable capability manifest
skills/stingray/references — capability mapping, alert DSL, north-star scenarios, examples, troubleshooting

Repository Layout

.
├── .claude-plugin/
│   └── marketplace.json
├── .github/
│   └── workflows/
│       └── validate.yml
├── AGENTS.md
├── CHANGELOG.md
├── CODEOWNERS
├── CONTRIBUTING.md
├── LICENSE
├── README.md
├── scripts/
│   └── validate-public.sh
└── skills/
    └── stingray/
        ├── LICENSE.txt
        ├── prompts.md
        ├── README.md
        ├── SKILL.md
        └── references/

Manual Install

Copy skills/stingray into your agent's configured skills directory.

Support

For packaging or distribution issues, contact [email protected].

License

Apache-2.0. See LICENSE.