The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill performs network requests to fetch updates and verify academic metadata. \n
scripts/paperspine_update.py downloads version manifests and skill archives from the author's GitHub repository (github.com/WUBING2023/PaperSpine) to perform self-updates. \n
scripts/citation_quality_audit.py and scripts/citation_verification_en.py connect to Crossref (api.crossref.org) and Semantic Scholar (api.semanticscholar.org) to resolve academic identifiers. \n- [COMMAND_EXECUTION]: Local shell commands are used for environment setup and workflow orchestration. \n
scripts/launch_paperspine_ui.ps1 executes PowerShell with -ExecutionPolicy Bypass to launch the interactive configuration wizard. \n
scripts/progress_check.py executes sibling Python scripts (such as artifact_check.py and integrity_audit.py) via subprocess.run to validate workflow stages. \n
The skill instructions require the use of system tools like pandoc and LaTeX compilers (pdflatex, xelatex) for document generation. \n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user documents, creating a potential attack surface for indirect prompt injection. \n
Ingestion points: Manuscript drafts and research materials are ingested via scripts/material_inventory.py and scripts/reference_inventory.py. \n
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used during data ingestion. \n
Capability inventory: The skill possesses the ability to execute Python scripts via subprocess.run (scripts/progress_check.py) and modify local configuration files (scripts/paperspine_update.py). \n
Sanitization: Content is validated for length and structure but not for malicious natural language instructions.

paper-spine