The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external web pages which could be used to deliver malicious instructions.
Ingestion points: Web content is ingested via context_dev_scrape_markdown and context_dev_extract_structured_data tools as defined in references/sandbase-api-map.md.
Boundary markers: Absent. The skill lacks explicit delimiters or instructions to ignore embedded commands within scraped content.
Capability inventory: The skill has the ability to read and write to a local TASKS.md file and invoke external scraping tools.
Sanitization: Absent. There is no evidence of sanitization or validation of the data retrieved from external sources before it is processed.
[COMMAND_EXECUTION]: The skill uses a dynamic tool-invocation pattern to execute scraping functionality.
It employs sandbase_describe_tool and sandbase_call_tool to interact with context_dev_scrape_markdown and context_dev_extract_structured_data based on runtime-resolved schemas.

task-management