Security audit
Security checks for vulnerabilities and agentic risk
This appears to be a legitimate crypto signal tool that contacts market/news APIs and stores local signal records, with no hidden destructive or exfiltration behavior found.
Install only if you are comfortable providing NewsAPI and optional CryptoPanic/Binance keys, sending market/news queries to those services or a configured proxy, and keeping generated trading-signal history on disk. Treat its BUY/SELL/HOLD outputs as informational analysis, not automatic trading authority or financial advice.
The skill advertises capabilities that include environment access, file read/write, network access, and shell execution, but it does not declare any explicit tool scope or permissions boundaries. This increases the chance the agent invokes powerful actions implicitly, making credential access, local file modification, or external requests possible without clear least-privilege constraints.
The trigger phrases are broad enough to match normal conversation about buying, selling, macro news, or crypto generally, which can cause the skill to activate unintentionally. In this skill, accidental activation is more concerning because activation may lead to shell, network, and file operations rather than a harmless text-only response.
The skill instructs the agent to write an llm_judgment field back into a signal JSON file and then perform later verification, creating persistent state across runs. Persisting LLM-generated content in local files can taint future processing, create an audit trail containing untrusted text, and enable unintended cross-session influence if later steps trust or reuse that field.
## Rules
- NEVER print raw JSON or file contents; summarize key fields only.
- NEVER invent probability forecasts or advice beyond what the script outputs.
- After `signal` completes: write `llm_judgment` into the signal JSON (1-2 sentences max), then run `verify`. Do NOT do anything else.
- When reading any `.json` file, extract only: direction, confidence, score, key factor votes, news_summary. Discard the rest.
## Setup
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
- When reading any `.json` file, extract only: direction, confidence, score, key factor votes, news_summary. Discard the rest.
## Setup
Credentials: `~/.openclaw/credentials/crypto-signal.json` (auto-created, chmod 600).
**Required:** `News_apiKey` / `NEWSAPI_KEY` · **Optional:** `CP_apiKey`/`CRYPTOPANIC_KEY`, `Bian_apiKey`/`BINANCE_API_KEY`, `proxy`/`HTTPS_PROXY` (mainland China: `http://127.0.0.1:7890`)
## Usage
The skill explicitly references external APIs, proxy support, and a local credential file, but it does not clearly warn the user that data may be transmitted to third parties or that local secrets will be accessed. This creates a transparency and consent problem, especially in an agent setting where users may not expect market symbols, prompts, or derived data to be sent externally.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
_SIGNALS = _ROOT / "signals"
_RECORDS = _ROOT / "records"
BINANCE_BASE = "https://api.binance.com/api/v3"
CRYPTOPANIC_BASE = "https://cryptopanic.com/api/v1/posts/"
NEWSAPI_BASE = "https://newsapi.org/v2/everything"
The request parameters force language: "en", which imposes a specific language policy on users regardless of their locale or preferences. The file does not offer an opt-in/override mechanism or explain why English-only filtering is required.
This code writes generated trading signals to disk, and elsewhere in the file similar writes persist verification results and historical records. While the writes are part of the program's functionality, there is no explicit user-facing warning here that running the tool stores market, signal, and verification data locally across multiple directories.
No suspicious patterns detected.