Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/nytc69/review-loop/executenpx skills add NYTC69/review-loop --skill executegit clone --depth 1 https://github.com/NYTC69/review-loopWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00096 | $0.08721 |
| Opus 5 | $0.00048 | $0.04360 |
| Sonnet 5 | $0.00019 | $0.01744 |
| Haiku 4.5 | $0.00010 | $0.00872 |
Grade A, and why
execute scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
the scheduler via `subprocess.PIPE` and surfaced through each How it starts
The opening of the file, as written. The whole thing — 727 lines — stays where its author put it; the contents beside it link to each section on GitHub.
execute — Codex Stage 1 Execution Sub-Skill
Drive an approved / user-supplied / review-only plan through the
execution loop, quality polish (Step 3.5), docs consistency (3.6),
security preflight (3.7), and delivery (Step 4). Supports multi-batch
runs via --stop-after <stage> and the unsafe --accept-external-state
opt-in.
Stage 1 Scope
- Codex Stage 1 follows the same broad
exec -> polish -> docs -> security -> deliverylifecycle. - Codex Stage 1 assumes a single orchestrator-owned workspace for the session.
- Included: execution loop, Quality Polish (Step 3.5), Documentation
Consistency (3.6), Security Preflight (3.7), Delivery (Step 4),
shared
.review-loop/config.md, shared.review-loop/sessions/*.md, Claude CLI default reviewer, optional local Codex reviewer, shared reviewer schema, Stage 1 hallucination guards. - Excluded: planning-phase orchestration — that lives in
.agents/skills/plan/SKILL.md.
Protocol Imports
The Orchestrator MUST Read each of these files at start. They are the single source of truth for this skill's execution loop and output schemas.
docs/protocol/session-file.mddocs/protocol/execution.mddocs/protocol/executor-output.mddocs/protocol/reviewer-output.md
Do not re-derive any rule that already lives in a protocol doc. When a
step below says "see docs/protocol/<doc>.md §Foo", follow that doc
verbatim. The startup read set is complete only after all 4 docs above
have been read explicitly; embedded executor/reviewer prompt bodies are
not a substitute for reading executor-output.md and
reviewer-output.md.
Runtime Identity
- Codex is the orchestrator. Do not do the planning or coding in the main thread.
- The orchestrator is the only writer of
.review-loop/sessions/{uuid}.md. review_loop_executornever writes the session file directly.review_loop_reviewernever writes the session file directly.- Do not create or switch to another git worktree or repository checkout.
- The Executor must stay in the orchestrator-owned workspace for the session. Executor-created hidden worktrees are forbidden in Codex Stage 1.
- When invoking Codex subagents, use a fresh self-contained prompt that embeds the required task context directly. Do not rely on inherited or forked parent thread context.
- Reject malformed Executor or Reviewer output instead of guessing.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 727 lines · 96 tokens per session scan A 8c412018049a
execute is a skill published in the GitHub repository NYTC69/review-loop (3 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 96 tokens to every session and 8,721 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
codex-adversarial-review-lite
Codex Adversarial Review - Lite: user-invoked audit workflow for Claude Code users who want Codex CLI to independently review AI-generated code, plans, test expectations, and scope before fixes are applied. Cross-platform (Windows, macOS, Linux, WSL). Use only when the user explicitly invokes audit or selftest.
codex-handoff
Three-phase collaboration workflow between Claude Code (planner/decision-maker) and Codex CLI (implementer/reviewer) via the codex-plugin-cc plugin. Use whenever a coding task involves multiple files, new modules, cross-cutting refactors, business logic changes, or anything that warrants a written spec and a review…
brain-docs
Set up and maintain a "project brain" — a brain/ folder holding three living documents (decisions.md, flow.md, todo.md) that give a project persistent memory across chat sessions. Use this skill whenever the user wants a decision log, architecture/flow notes, project memory, living documentation, or notes that survive…
codex-handoff
Three-phase collaboration workflow between Claude Code (planner/decision-maker) and Codex CLI (implementer/reviewer) via the codex-plugin-cc plugin. Use whenever a coding task involves multiple files, new modules, cross-cutting refactors, business logic changes, or anything that warrants a written spec and a review…
moai-ref-ui-polish
UI polish and interface-completion reference: the small visual details — concentric border radius, optical alignment, shadow-vs-border, motion easing, typography smoothing, tabular numbers, icon stroke weight, hit areas — that separate polished interfaces from generic ones. Agent-extending skill that amplifies…
trellis-brainstorm
Guides collaborative requirements discovery before implementation. Creates task directory, seeds PRD, asks high-value questions one at a time, researches technical choices, and converges on MVP scope. Use when requirements are unclear, there are multiple valid approaches, or the user describes a new feature or complex…