The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for processing untrusted data (user-supplied classic DQL queries) to rewrite them into Smartscape DQL. This represents a potential vulnerability surface, though it is necessary for the skill's primary purpose.
Ingestion points: User-provided classic DQL queries (SKILL.md)
Boundary markers: None identified in the provided skill files
Capability inventory: Rewriting and generating DQL queries for execution (SKILL.md, references/migration-workflow.md)
Sanitization: No specific sanitization or validation of the input query strings is mentioned
[PROMPT_INJECTION]: The instructional content does not attempt to bypass safety filters, override agent behavior, or extract system prompts. The use of "IMPORTANT" and "CRITICAL" labels is limited to standard technical guidance.
[DATA_EXPOSURE]: No hardcoded credentials, sensitive file paths, or unauthorized network operations were detected in the instructions or code snippets.
[UNVERIFIABLE_DEPENDENCIES]: The skill does not include any requirements files, package installations, or remote script execution patterns. It suggests loading another skill from the same vendor (dt-dql-essentials) for syntax validation.
[OBFUSCATION]: All content is provided in plain text. No Base64 encoding, zero-width characters, or other hidden content techniques were found.

dt-migration