The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data through search, scraping, and enrichment capabilities via the sandbase_run tool.
Ingestion points: Data retrieved from external sources (scraping, search results) via the sandbase_run and sandbase_discover tools as described in SKILL.md and references/sandbase-api-map.md.
Boundary markers: The instructions do not specify the use of clear boundary markers (e.g., XML tags or specific delimiters) or explicit "ignore embedded instructions" warnings to isolate untrusted external content from the agent's internal instructions.
Capability inventory: The skill has the ability to execute network-based tool calls (sandbase_run, sandbase_discover) and interacts with the local environment to produce implementations, analysis, or artifacts as part of its workflow.
Sanitization: While the skill encourages translating requests into answerable questions, there are no explicit instructions for sanitizing, escaping, or filtering potential instructions that might be embedded in the retrieved external data before it is processed by the agent.

database-lookup