The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes curl to retrieve blockchain and market data from mempool.space and coinbase.com, and uses bun run to execute local scripts within the skill's environment. These commands are necessary for the skill's functionality and target reputable sources.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from news signals which poses a vulnerability surface for instruction injection.
Ingestion points: Untrusted content is ingested via the news_signals tool as defined in SKILL.md.
Boundary markers: The skill does not employ delimiters or explicit instructions to the agent to ignore content-embedded commands.
Capability inventory: The agent can perform write operations to the platform, specifically filing corrections and reports via the news_correct_signal and news_file_signal tools.
Sanitization: No sanitization or filtering of the ingested signal content is specified in the skill instructions.

aibtc-news-fact-checker