The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's instructions in SKILL.md direct the agent to execute local Python scripts, specifically scripts/calculate_scores.py and scripts/generate_schematic.py, using the Bash tool to perform score aggregation and generate publication-quality diagrams.- [INDIRECT_PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection due to its document processing nature.
Ingestion points: The skill is designed to ingest and analyze untrusted external scholarly work (research papers, proposals, literature reviews) during the evaluation workflow described in SKILL.md.
Boundary markers: The instructions do not define explicit delimiters or boundary markers to isolate untrusted content when it is interpolated into the agent's context.
Capability inventory: The skill has access to the Bash (command execution) and Write (file system modification) tools, as specified in the frontmatter of SKILL.md.
Sanitization: There are no documented procedures for sanitizing or filtering potential malicious instructions embedded within the scholarly documents being evaluated.

scholar-evaluation