The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes various commands using sealos-cli and a local Node.js script.
Evidence: SKILL.md instructs the agent to run node <SKILL_DIR>/scripts/analyze-project-s3.mjs and various sealos-cli s3 commands for bucket and credential management.
[EXTERNAL_DOWNLOADS]: The skill may download and execute the sealos-cli tool at runtime.
Evidence: SKILL.md suggests using npx -y sealos-cli@latest ... for one-off commands if the binary is not found locally. This fetches the package from the NPM registry.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it analyzes untrusted project files.
Ingestion points: scripts/analyze-project-s3.mjs scans and reads content from arbitrary project files (e.g., .env, source code, configuration files).
Boundary markers: Absent. The instructions do not specify delimiters or warnings to ignore instructions embedded within the analyzed project files.
Capability inventory: The skill can execute shell commands via sealos-cli and modify local .env files.
Sanitization: None. The analyzer script extracts raw text patterns from files to determine storage requirements.

sealos-s3