Security risk report for @forgecat/openai_skills_yeet v0.0.11
Source Integrity
Low
Profile is attributed to OpenAI's official skills repository with a specific commit hash and Apache-2.0 license.
Metadata (author, original repository, version, commit date) is consistent with a legitimate upstream source.
Agent Intent
Low
The skill describes a legitimate Git/GitHub workflow (stage, commit, push, PR) with no instructions to manipulate the AI, exfiltrate data, or bypass security controls.
Naming conventions and workflow steps are transparent and focused on the stated purpose; no hidden instructions or prompt injection attempts detected.
The skill does not instruct reading credentials, installing remote payloads, or revealing system prompts.
Details
Evidence
Workflow section contains only standard Git commands and GitHub CLI operations.
Prerequisites explicitly require user authentication checks before proceeding.
No references to environment variables, credential files, or external exfiltration endpoints.
Permissions
PRM-000Medium
The skill executes shell commands (`git`, `gh`) that can modify the repository and push to remote, which is broader than read-only operations.
Authority is justified by the stated purpose (Git workflow automation), but shell execution carries inherent risk if misused.
Details
Evidence
Commands include `git add -A`, `git commit`, `git push -u origin`, and `gh pr create` — all state-mutating operations.
The skill operates on the user's repository and GitHub account, requiring trust in the workflow logic.
MCP Risk
Low
No MCP servers are declared in this profile.
The skill relies only on standard CLI tools (`git`, `gh`) that are user-installed and authenticated.
Details
Evidence
MCP servers section is empty.
Dependencies list is empty; no external binaries or network services are invoked.
ForgeCat