The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to use npx @useclawlink/cli, which downloads and executes code from the NPM registry at runtime.
Evidence: Multiple instructions in SKILL.md call npx @useclawlink/cli to login, connect, and run actions.
Note: The NPM package scope @useclawlink differs from the author's identified vendor name clawlink-hq, which typically uses the @clawlink-hq prefix.
[COMMAND_EXECUTION]: The skill is entirely reliant on the agent's ability to execute shell commands to interact with the Mural service.
[CREDENTIALS_UNSAFE]: The skill references a sensitive local credential store at ~/.clawlink/credentials.json where authorization tokens are maintained.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process data from Mural whiteboards, which represents a potential attack surface for indirect prompt injection if the ingested content contains malicious instructions.
Ingestion points: Data read from Mural via npx @useclawlink/cli actions mural as described in SKILL.md.
Boundary markers: Absent; there are no instructions to the agent to ignore embedded commands in the whiteboards.
Capability inventory: Shell command execution via npx (subprocess call) present in SKILL.md.
Sanitization: Absent; the instructions do not specify any filtering or validation of the retrieved whiteboard content.

mural