The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires the execution of shell commands to maintain the local ledger and interact with remote APIs.
Utilizes git add and git commit to track Markdown files within the repository structure.
Utilizes official command-line interfaces (gh for GitHub, glab for GitLab, tea for Gitea) to synchronize local issues with remote trackers.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a hand-off mechanism, creating documentation meant to be read and acted upon by other AI agents or in future sessions.
Ingestion points: Incorporates user-provided descriptions of feature gaps, technical debt, and architectural questions into persistent files.
Boundary markers: Employs a structured Markdown template with explicit YAML frontmatter and a standard ## Origin section to compartmentalize issue metadata.
Capability inventory: Has permissions for local file writing to llmwiki/wiki/sources/draft/, Git operations, and external issue publishing via CLI tools.
Sanitization: The skill instructions do not specify any sanitization or validation logic for the user-supplied content before it is written to the ledger files.

raise-issue