The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes profile attributes, event history, and segment memberships from the Lytics API and interpolates them directly into the agent's context.
Ingestion points: explore.md (Step 1) and investigate.md (Step 1) fetch user-controlled profile data.
Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands in the processed data.
Capability inventory: The skill has the ability to execute DELETE operations on profiles via curl in lookup.md.
Sanitization: No sanitization or validation of the fetched profile data is performed before presentation.
[COMMAND_EXECUTION]: The skill instructs the agent to execute curl commands for data retrieval and profile deletion. While it implements a confirmation gate for mutations in references/confirmation-gate.md, the agent retains the capability to perform irreversible deletions.
[DATA_EXFILTRATION]: The skill accesses ~/.lytics/accounts.toml to retrieve API tokens for different accounts. This is a sensitive configuration file containing credentials. As this is a vendor-owned resource for 'lytics', this access is documented as standard functionality for multi-account management.

lytics-profiles