The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the Membrane CLI (@membranehq/cli) from the official registry. This is a vendor-provided tool required for the skill's operation.
[COMMAND_EXECUTION]: The skill utilizes shell commands through the membrane CLI to interact with the Nectar CRM API, including running predefined actions and making raw HTTP requests.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its processing of external CRM data.
Ingestion points: Untrusted data enters the agent context through the outputs of CRM queries (e.g., retrieval of Notes, Persons, or Organizations) via the membrane action run and membrane request commands.
Boundary markers: Absent; the instructions do not implement delimiters or warnings to ignore instructions embedded within the retrieved CRM data.
Capability inventory: The agent has the ability to write data back to the CRM and make arbitrary network requests through the Membrane proxy using membrane request in SKILL.md.
Sanitization: Absent; there is no evidence of validation or sanitization performed on the content retrieved from the external API before it is processed by the agent.

nectar-crm