Skill Claude CodeCodex
Security engineer focused on vulnerability detection, threat modeling, and secure coding practices. Use for security-focused code review, threat analysis, or hardening recommendations.
Skill Claude CodeCodex
Security engineer focused on vulnerability detection, threat modeling, and secure coding practices. Use for security-focused code review, threat analysis, or hardening recommendations.
Skill Claude CodeCodex
QA engineer specialized in test strategy, test writing, and coverage analysis. Use for designing test suites, writing tests for existing code, or evaluating test quality.
Command
Run the full engineering cycle autonomously end-to-end — spec → plan → build → test → review → ship — making reasonable calls without stopping at every gate. Use /full-cycle instead when the user wants to approve each gate; use /goal instead when there's a measurable objective to converge on.
Command
Search the Hamzaish brain — learnings, decisions, playbooks, product docs. Use for "what did we decide about X" / "have we seen this before" recall, or before re-deriving something. Returns ranked citations with snippets.
Command
Refresh Hamzaish's brain index — re-scan all known folders and update SQLite FTS5 store.
Command
Implement the next task incrementally — build, test, verify, commit.
Command
hamzaish.md.
Command
Manual save-point commit with a real message. Use when you want a named milestone in the log between auto-commits.
Command
Simplify code for clarity and maintainability — reduce complexity without changing behavior.
Command
Orchestrate the full engineering cycle — setup → spec → plan → test → build → review → ship — pausing at each gate for approval. Use /auto instead when the user says go/just-do-it (no gates); use /goal instead when converging on a measurable objective.
Command
Pursue a measurable objective autonomously — score against a rubric, fix the biggest gap, verify, log every run, and stop only when a fresh-eyes agent confirms the goal is met (or the run budget is spent). Use /auto instead for a feature build with no numeric target; use /write-a-goal first if the objective isn't…
Command
Momentum router. Default is BUILD — routes straight to the full-cycle build engine. Strategy rails (idea/pricing/GTM) and stage-resume are opt-in side doors, never a toll. Skip is available at every step.
Command
Run the impact-scored learning loop at a major-cycle boundary — gather the cycle's candidate learnings, score them with the rubric, promote the top few into guardrails, and append a changelog line.
Command
Break work into small verifiable tasks with acceptance criteria and dependency ordering.
Plugin Claude Code
Bundles 1 skill, 7 commands, 1 hook · 193 tokens together
Claude writes. Codex reviews. You ship. Cross-model pair programming inside Claude Code.
Command
Snapshot all products — table + top 3 priorities + on-fire + don't-touch. The "where should I focus today" command.
Command
One-command repo ship — branch → commit → PR → wait for CI → squash-merge → sync local. The safe GitHub flow, hidden behind one step, narrated in plain English. For shipping a change to a repo's main; NOT a product production deploy (that's /ship).
Command
Review the auto-captured learning queue and hand survivors to the brain (distilled, never pasted). The human gate between automatic capture and /learn-loop.
Command
Cut the next public release at a major-cycle boundary — pick the next semver tag, assemble notes from meta/changelog.md, create an annotated tag + GitHub Release, and mark it latest.
Command
Conduct a five-axis code review — correctness, readability, architecture, security, performance.
Command
Fast security baseline audit of a product — tracked secrets, unpinned/vulnerable GitHub Actions, workflow permission scope, untrusted-input triggers, MCP-config surface, and an RLS reminder. Add --live to also check the RUNNING system for drift (live RLS status, storage-bucket visibility, Security Advisors, preview…
Command
Scaffold or refresh a project's Claude Code file structure — CLAUDE.md, path-scoped rules, docs, gitignored personal overrides, and a starter /ship command.
Command
Promote reviewed commit(s) from the working branch to the product's production branch and push — the single deploy action. Auto-commit wip snapshots stay on the working branch and never reach production. For deploying a PRODUCT to production; use /pr instead when shipping a change to a repo's main branch.
Command
Start spec-driven development — write a structured specification before writing code.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: