The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through the ingestion of untrusted data from the Nostr network.
Ingestion points: Commands read-feed, search-tags, and get-profile fetch content (notes and profiles) from public Nostr relays into the agent's context.
Boundary markers: The output does not include boundary markers or instructions to the agent to ignore embedded commands within the fetched data.
Capability inventory: The agent has the ability to write back to the network (via post and amplify-signal) and access wallet keys, creating a risk if fetched content contains malicious instructions that the agent executes.
Sanitization: No content filtering or sanitization is performed on the data retrieved from relays.
[EXTERNAL_DOWNLOADS]: The amplify-signal command fetches data from https://1btc-news-api.p-d07.workers.dev/takes/. This is a vendor-specific resource associated with aibtc.news used for its intended purpose of signal amplification.
[DATA_EXPOSURE]: The skill uses the agent's Bitcoin BIP84 derivation path (m/84'/0'/0'/0/0) to derive its Nostr identity. As noted in the documentation, this creates a shared risk where a compromise of the Nostr signing environment or keys directly impacts the security of the associated Bitcoin wallet.
[COMMAND_EXECUTION]: The skill implements a CLI-based execution model where the agent can trigger cryptographic signing and network broadcasts to public relays.

nostr