The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of CLI tools including sf, curl, and jq. It uses shell redirection to save command outputs to a temporary scratch directory for processing. The instructions guide the agent to construct complex commands and SOQL queries using these tools.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection through its ingestion of untrusted configuration data.
Ingestion points: User-supplied inputs such as {PLATFORM_KEY}, {LINE_ACCESS_TOKEN}, {WABA_ID}, and {CHANNEL_NAME} which are interpolated into shell commands and SOQL query strings in SKILL.md and the reference files.
Boundary markers: None. The skill does not provide delimiters or instructions to the agent to treat these inputs as untrusted data strings.
Capability inventory: The skill has the capability to execute shell commands, perform network operations via sf and curl, and write/read files in the local filesystem.
Sanitization: Absent. There are no instructions to sanitize, escape, or validate the content of the variables before they are interpolated into executable shell contexts.
[EXTERNAL_DOWNLOADS]: The skill uses curl in references/line.md to communicate with the official LINE API at api.line.me. This is a well-known service required for the skill's primary purpose of verifying LINE messaging tokens.

service-de-channel-create