The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the 'openkakao-rs' CLI to perform messaging tasks and includes a --hook-cmd flag that allows the execution of arbitrary shell commands in response to incoming message events.\n- [EXTERNAL_DOWNLOADS]: Installs the 'openkakao-rs' binary from the author's official Homebrew tap (JungHoonGhae/openkakao).\n- [DATA_EXFILTRATION]: Provides functionality to send real-time chat data to external endpoints via the --webhook-url flag, which can be used to transmit message content outside the local environment.\n- [CREDENTIALS_UNSAFE]: Includes commands like login --save designed to extract authentication tokens from the local KakaoTalk application's Cache.db file for session management.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted chat data from the KakaoTalk platform.\n
Ingestion points: Incoming chat messages are ingested via the watch and loco-read commands as described in SKILL.md.\n
Boundary markers: Absent; the instructions do not provide delimiters or warnings to ignore instructions embedded in chat content.\n
Capability inventory: The skill can execute local commands via --hook-cmd and perform network operations via --webhook-url.\n
Sanitization: Absent; no validation or cleaning of ingested message content is specified.

openkakao-cli