The Agent Skills Directory

[PROMPT_INJECTION]: No attempts to override agent safety guidelines, bypass instructions, or extract system prompts were detected. The skill uses instructional language to enforce a specific workflow (TDD-first orchestration) without malicious intent.
[DATA_EXFILTRATION]: No sensitive file paths, credential markers, or unauthorized network operations were identified. The skill does not attempt to access local environment variables or private configuration files.
[REMOTE_CODE_EXECUTION]: The skill does not download external scripts or packages. It references local sub-skills via relative paths in the assets/skill-map.json file for internal routing purposes.
[COMMAND_EXECUTION]: No shell command execution or subprocess spawning patterns were found in the skill instructions or the associated JSON mapping file.
[OBFUSCATION]: The skill files are written in plain text and standard JSON format. No Base64, hex encoding, zero-width characters, or homoglyph substitutions were detected.
[INDIRECT_PROMPT_INJECTION]: As an entry-point orchestrator, the skill ingests user requests to decompose them into tasks. While this represents a standard attack surface for indirect prompt injection, the skill lacks autonomous dangerous capabilities and functions purely as a logic router within the Elixir/Phoenix ecosystem.
[PRIVILEGE_ESCALATION]: No commands related to privilege escalation (e.g., sudo, chmod) or persistent mechanism installation were found.
[DYNAMIC_EXECUTION]: The skill does not generate or compile code at runtime. It serves as a persona that provides instructions to the agent on how to proceed with development tasks.

elixir-skill-router