The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process library documentation from external sources via the mcp__context7__query-docs tool. This content is untrusted and could potentially contain instructions intended to manipulate the agent's behavior, particularly when synthesizing findings or generating reports.
Ingestion points: Documentation content retrieved through mcp__context7__query-docs as described in SKILL.md.
Boundary markers: The workflow lacks explicit delimiters or instructions to ignore potential commands embedded within the retrieved documentation snippets.
Capability inventory: The skill utilizes a local Python script (scripts/new_report.py) to generate research reports, which involves file system write operations.
Sanitization: There is no evidence of sanitization or filtering of external documentation content before it is processed or used as input for the report generation script.
[COMMAND_EXECUTION]: The skill instructions in SKILL.md direct the agent to execute a local script (python3 scripts/new_report.py) to automate the creation of research reports. The script is a template-based markdown generator that uses standard library modules (argparse, pathlib, datetime). While the script itself is benign, the practice of executing local code based on externally-sourced data presents a vulnerability surface if the execution environment does not properly isolate the process or sanitize the arguments.

context7-research