The Agent Skills Directory

[DATA_EXPOSURE]: The script scripts/correlation_tracker.py accesses a local SQLite database at ~/.polymarket-paper/portfolio.db to retrieve trade history and holdings. While this is the intended data source for the skill's correlation analysis feature, it involves reading potentially sensitive personal financial activity from the local filesystem.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the Polymarket API and local database which can contain attacker-controlled text in market descriptions.
Ingestion points: scripts/find_edges.py, scripts/momentum_scanner.py, and scripts/correlation_tracker.py fetch and process market_question and description fields from external APIs (gamma-api.polymarket.com) and the local portfolio database.
Boundary markers: None. The skill does not use delimiters or instructions to the agent to ignore embedded commands within market data.
Capability inventory: The skill provides data to the agent context but does not directly execute subprocesses or dynamic code; however, the agent's interpretation of the extracted text could lead to malicious actions if it encounters injected instructions.
Sanitization: The scripts use regex for keyword extraction but do not sanitize or escape the content for prompt safety before presenting it to the agent context.

polymarket-analyzer