From codex-advisor
Refine a rough coding request into a precise spec using current repository context before Claude Code executes it.
How this skill is triggered — by the user, by Claude, or both
Slash command
/codex-advisor:refine-with-contextThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use this skill when the user explicitly invokes `/codex-advisor:refine-with-context` or asks to refine a request using current repository context.
Use this skill when the user explicitly invokes /codex-advisor:refine-with-context or asks to refine a request using current repository context.
Your job is to call the Codex Advisor bridge, show the refined spec, and stop for user confirmation. Do not implement the refined spec unless the user explicitly confirms.
Pass the request on stdin so shell quoting cannot change it:
node "${CLAUDE_PLUGIN_ROOT}/scripts/codex-refine.mjs" --with-context <<'CODEX_ADVISOR_REQUEST'
<request to refine>
CODEX_ADVISOR_REQUEST
The bridge gathers a bounded file tree, current git status and diff, and ripgrep matches for relevant request terms before starting the Codex app-server turn.
npx claudepluginhub saadjs/codex-advisor --plugin codex-advisorProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.