Security risk report for @forgecat/openai_skills_pdf v0.0.11
Source Integrity
Low
Profile sourced from OpenAI's official skills repository (https://github.com/openai/skills) with clear attribution and commit hash.
Licensed under Apache-2.0; no supply-chain red flags or untrusted third-party dependencies.
Agent Intent
Low
Content describes legitimate PDF processing workflows using well-known, standard Python libraries (reportlab, pdfplumber, pypdf) and system tools (Poppler/pdftoppm).
No instructions to read credentials, exfiltrate data, ignore system prompts, or execute remote payloads.
Guidance on dependency installation and rendering is standard practice documentation, not poisoning; no typosquatted or backdoored packages recommended.
Details
Evidence
"Use `reportlab` to generate PDFs when creating new documents."
"Use `pdfplumber` (or `pypdf`) for text extraction and quick checks"
"Prefer `uv` for dependency management." — standard tooling recommendation
Permissions
Low
No tool definitions or MCP servers declared; skill is purely instructional guidance for the agent.
Recommended system tools (pdftoppm, Poppler) and Python packages are scoped to PDF processing, matching the stated purpose.
No file_write, file_delete, shell, or network permissions requested beyond what is necessary for PDF workflows.
Details
Evidence
"(none declared)" — no tools or MCP servers in the profile.
"Use `tmp/pdfs/` for intermediate files; delete when done." — scoped, temporary file usage.
MCP Risk
Low
No MCP servers defined in the profile.
No hidden instructions, arbitrary binary execution, or unrestricted network/filesystem access.
ForgeCat