The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches data from several established cryptocurrency data providers and protocol APIs.
Evidence: bridge_fetcher.py, protocol_adapters.py, and settings.yaml reference endpoints for DefiLlama (bridges.llama.fi), Wormhole (api.wormholescan.io), LayerZero (scan.layerzero.network), and Across (across.to).
[COMMAND_EXECUTION]: The skill executes local Python scripts to aggregate and process bridge data. The execution is restricted to the skill's own monitor scripts.
Evidence: SKILL.md limits the Bash tool to the pattern python:*bridge_monitor*, which corresponds to the provided bridge_monitor.py CLI utility.
[DATA_EXFILTRATION]: Network activity is limited to the defined monitoring sources. There is no evidence of the skill attempting to access or transmit sensitive system files or credentials to external servers.
[CREDENTIALS_UNSAFE]: The documentation includes guidance on managing API keys through environment variables or separate configuration files (config/crypto-apis.env), which align with standard security practices for secret management.
[PROMPT_INJECTION]: The instructions in SKILL.md are focused on task execution and do not contain patterns intended to bypass safety guidelines or override agent behavior.

monitoring-cross-chain-bridges