The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external CI logs, which could contain malicious instructions intended to mislead the agent. \n
Ingestion points: The skill fetches logs from GitHub Actions, GitLab CI, Azure Pipelines, and Bitbucket Pipelines using CLI tools and APIs. \n
Capability inventory: Access to shell execution for Git operations and file system writing. \n
Boundary markers: No delimiters or protective instructions are used to isolate untrusted log content during parsing. \n
Sanitization: Log output is parsed for specific markers but not validated or sanitized against prompt injection patterns. \n- [COMMAND_EXECUTION]: The skill executes sensitive Git commands that alter repository history, such as amending commits and force-pushing. \n
Evidence: Instructions for git commit --amend and git push --force-with-lease are present in SKILL.md. \n- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and execute setup scripts from external sources. \n
Evidence: container-engine.md includes curl -fsSL https://get.docker.com | sudo sh. \n- [PRIVILEGE_ESCALATION]: The skill requests administrative privileges to install container engines and manage system services. \n
Evidence: Use of sudo apt-get install, sudo dnf install, and sudo systemctl in container-engine.md. \n- [REMOTE_CODE_EXECUTION]: The skill dynamically executes remote code through package managers and scripts. \n
Evidence: Use of npx mega-linter-runner for local linting and curl | sh for Docker installation.

megalinter-check