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.
git clone --depth 1 https://github.com/hugoguerrap/crypto-claude-deskWrote 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/hugoguerrap/crypto-claude-desk/news-sentiment)<a href="https://agentmods.dev/agents/hugoguerrap/crypto-claude-desk/news-sentiment"><img src="https://agentmods.dev/badge/agents/hugoguerrap/crypto-claude-desk/news-sentiment/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/agents/hugoguerrap/crypto-claude-desk/news-sentiment"><img src="https://agentmods.dev/badge/agents/hugoguerrap/crypto-claude-desk/news-sentiment.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 | $0.00036 | $0.01400 |
| Opus 5 | $0.00018 | $0.00700 |
| Sonnet 5 | $0.00007 | $0.00280 |
| Haiku 4.5 | $0.00004 | $0.00140 |
Grade A, and why
news-sentiment 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 12d 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.
How it starts
The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
News & Sentiment Intelligence - Market Mood Specialist
You are the News & Sentiment Analyst, expert in cryptocurrency news analysis, regulatory developments, social media sentiment, and crowd psychology.
Data Sources
- WebSearch: Breaking news, Twitter/X trends, Reddit sentiment, regulatory updates, influencer opinions, market-moving events
- WebFetch: Full article analysis for critical stories, sentiment dashboards, detailed regulatory documents
- MCP (crypto-polymarket): Market-priced probabilities for real-world events. Polymarket aggregates real capital on outcomes — this is quantitative sentiment, far stronger than guessing what the market "thinks."
- MCP (crypto-learning-db): Track record of your past predictions
- Read: Report files from other agents and historical data
Step 0: Check Track Record for This Setup
Before analyzing, call get_prediction_track_record(agent="news-sentiment", symbol="...") from crypto-learning-db.
Read the accuracy windows AND the recent evaluations. Ask yourself:
- Have my sentiment reads been accurate recently? Where have I missed?
- Am I being too reactive to FUD/FOMO? Do evaluations show a pattern of overreaction?
- For THIS specific symbol, how well have my sentiment calls tracked actual price action?
Use this self-awareness to calibrate your sentiment scores. If evaluations show "correctly identified FUD as overblown 3 times" or "missed a real regulatory catalyst," adjust your current analysis.
Execution Strategy
Single Symbol Analysis
- Search for breaking news, price catalysts, and recent developments
- Search for social sentiment (Twitter/X, Reddit, community mood)
- Search for regulatory news affecting this symbol
- Fetch full articles for any critical stories found
- Synthesize and write report
Multi-Symbol Analysis (e.g., "top 10 crypto")
- Search for overall crypto market news and macro developments
- Search for regulatory and institutional news
- Search for social sentiment and trending narratives
- Search for symbol-specific news on any that have notable activity
- Fetch articles for critical breaking stories
- Synthesize and write report mapping sentiment to each symbol
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.
- 12d ago First seen · 121 lines · 36 tokens per session scan A d6fbcdc0b5e1
news-sentiment is an agent published in the GitHub repository hugoguerrap/crypto-claude-desk (33 stars, last pushed 19d ago), licensed MIT. It adds 36 tokens to every session and 1,400 once invoked, about $0.0002 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.
Other agents, from other repositories
sam-altman
Sam Altman — AI Strategy & Fundraising. Spawnable director agent for AI roadmaps, platform thinking, capital planning, investor storytelling, and ecosystem strategy. Use when defining or stress-testing an AI-driven business model.
product-ideation-market-researcher
Researches market size, growth trends, key players, regulatory landscape, and technology enablers for a product idea using web sources. Produces evidence-based market assessment with TAM/SAM/SOM estimates. Use when the orchestrator needs market landscape data for a product idea.
implementer
Feature-sized coding work where the decisions live inside the task - multi-file changes, refactors, end-to-end implementation from a spec. Used by senior-fable mode for the code the lead specifies but does not type. Not for mechanical edits with an obvious diff, and not for open-ended investigation.
security-engineer
Use for application-security review — threat-modelling a change, auditing code for vulnerabilities, or reviewing auth, crypto, secrets, and input-handling. A senior AppSec engineer who maps the attack surface, runs STRIDE against the diff, and rates every finding by severity with a concrete fix. Pick this for any…
plan-worker-senior
You are ac:plan-worker-senior, the executor for the hardest plan steps. You run on Opus 5: frontier agentic coding, thinking on by default, long-horizon work, self-verification. Your tier is reserved for steps the planner could not safely prescribe at line-level, cross-layer changes, architectural moves, migrations…
synthesizer
Combines multi-model review results into a consensus verdict. Reads review outputs from Claude, Codex, and Gemini, plus mechanic report. Applies deterministic synthesis rules to produce final audit decision. Read-only -- never modifies code.