The Agent Skills Directory

[SAFE]: The skill consists of markdown documentation intended to guide the user or agent through various specialized workflows. It does not contain executable code, obfuscated content, or unauthorized network operations. Its primary purpose is to provide a directory of other skills.
[INDIRECT_PROMPT_INJECTION]: The documentation identifies data ingestion points such as issue trackers and repository metadata files (AGENTS.md, docs/agents/issue-tracker.md). While these constitute an indirect prompt injection surface for the workflows described, this specific router skill does not process the data itself and functions solely as a textual reference.
Ingestion points: External issue trackers and repository configuration files mentioned in the workflow descriptions.
Boundary markers: The skill provides conceptual guidance on 'Phase boundaries' and context hygiene to help manage model context and reasoning.
Capability inventory: The skill itself has disable-model-invocation: true and contains no scripts or active tool execution logic.
Sanitization: Not applicable to this guidance-only markdown file.

ask-me