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 commands/hmj1026/dhpk/codex-reviewgit clone --depth 1 https://github.com/hmj1026/dhpkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/hmj1026/dhpk/codex-review)<a href="https://agentmods.dev/commands/hmj1026/dhpk/codex-review"><img src="https://agentmods.dev/badge/commands/hmj1026/dhpk/codex-review.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.00020 | $0.00463 |
| Opus 5 | $0.00010 | $0.00231 |
| Sonnet 5 | $0.00004 | $0.00093 |
| Haiku 4.5 | $0.00002 | $0.00046 |
Grade A, and why
codex-review scanned grade A with 0 findings 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 4d ago.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
/codex-review
Use one Codex-backed review entrypoint. --scope selects the review contract:
| Scope | Review target | Preferred depth |
|---|---|---|
diff |
current uncommitted diff | fast for quick feedback, full before handoff |
branch |
<base>..HEAD feature branch |
full |
doc |
an explicit or changed document | full |
security |
changed security-sensitive code | full |
tests |
tests and their production seam | full |
--depth fast|full defaults to full. --coverage is valid only with
--scope tests; --spec is valid only with --scope doc. --base chooses a
branch comparison where relevant, and --continue resumes the Codex thread.
Read the matching prompt/reference from dhpk-change-review, dhpk-doc-review,
dhpk-security-review, or dhpk-test-review before calling Codex. Collect only
the selected scope, ask Codex for file:line findings and a verdict, and report
the executed local checks plus PASS/FAIL/SKIP evidence. A standalone security,
documentation, or test audit should prefer its dedicated reviewer/skill; these
scope modes are the consolidated CLI second-opinion path.
Do not auto-loop beyond the policy ceiling. A BLOCK verdict names the focused
fix and rerun command; an acceptable verdict records scope, evidence, and
remaining uncertainty.
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.
- 4d ago First seen · 35 lines · 20 tokens per session scan A 38e9041333b7
codex-review is a command published in the GitHub repository hmj1026/dhpk (2 stars, last pushed 4d ago), licensed MIT. It adds 20 tokens to every session and 463 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
init
在既有專案初始化 Octopus——摸專案、檢查 OpenSpec CLI 與結構（缺則引導安裝）、盤點 change 狀態、建 Arena、出交接報告。每個專案跑一次。.
harness-lint
이 프로젝트의 생성된 하네스를 계약에 대고 검사한다. 결정적 검사 9종.
harness-evolve
하네스를 쓴 결과를 회고하고 에이전트·스킬·오케스트레이터에 되먹인다.
build
SDD 後半段（全自主）：驗 Locked → Builder 逐 task 實作（每條 task 隨行回報）→ 審查 → P1 自動退修（≤3 輪）→ 帶驗收報告回來，你只剩 merge 與 archive。中途不停下來等人。.
spec
SDD 前半段：需求釐清（反問+挑戰）→ OpenSpec change 一次落檔（proposal＋delta＋tasks，Draft）→ TPM 拍板鎖定（回一個 OK 即可）。鎖定後 build 全自主執行。.
debug
用 Agent 工具啟動 debugger agent，把錯誤資訊原文（含截圖/log）完整轉給它。.