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 skills add ChrisLamDev/hermes-core-skills --skill agent-capability-comparison-methodologygit clone --depth 1 https://github.com/ChrisLamDev/hermes-core-skillsWrote 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/skills/chrislamdev/hermes-core-skills/agent-capability-comparison-methodology)<a href="https://agentmods.dev/skills/chrislamdev/hermes-core-skills/agent-capability-comparison-methodology"><img src="https://agentmods.dev/badge/skills/chrislamdev/hermes-core-skills/agent-capability-comparison-methodology.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 5.1 | $0.00007 | $0.01319 |
| Opus 5 | $0.00003 | $0.00660 |
| Sonnet 5 | $0.00001 | $0.00264 |
| Haiku 4.5 | $0.00001 | $0.00132 |
Grade A, and why
agent-capability-comparison-methodology 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 8d 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.
How it starts
The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Agent 能力比較方法論
適用場景
用戶想比較兩個 AI agent/tool 嘅能力差異，例如「Hermes Agent vs OpenAI Codex」、「CUA vs Codex Desktop Control」。
核心原則
唔靠「聽講」，用證據說話。 三層驗證：
- 官方文件 — GitHub README、docs 目錄、官方網站
- 原始碼證據 — 直接睇 source code 確認功能存在與否
- 實測驗證 — 真係裝黎用、試下得唔得
比較流程
第一層：快速掃描（30 分鐘）
- 睇 GitHub README 了解定位
- 睇 docs/ 目錄了解功能列表
- 記錄 GitHub stats（stars、forks、commits、issues、branches、tags）
第二層：深入驗證（1-2 小時）
逐個功能查原始碼：
| 功能類別 | 要 check 嘅嘢 |
|---|---|
| 記憶系統 | source code 有冇 memory/memories 目錄？點樣跨 session？ |
| Cron/排程 | 有冇 scheduler/cron crate/module？ |
| 多平台支援 | 有冇 telegram/discord/slack module？ |
| 模型自由度 | model provider 係硬編碼定可配置？ |
| Computer Use | 有冇 screenshot/click/type_text 工具？ |
| MCP 整合 | 有冇 mcp server/client 實作？ |
| Skills | 係 static 定 dynamic learning？ |
| Desktop App | 係咪真係有 Electron/Tauri app？ |
檢查方法：
ls -d */睇目錄結構grep -r "keyword" --include="*.py" --include="*.rs" --include="*.ts"搜尋關鍵字- 睇 GitHub Issues 用 label 過濾（如
computer-use、memory、cron） - 睇 Discussions 嘅 feature requests
第三層：實測驗證（按需要）
- 真係裝黎用
- 試核心功能
- 紀錄成功/失敗嘅 case
關鍵：Always distinguish CLI vs Desktop App features
- Codex Desktop has Computer Use + Automations; Codex CLI does NOT
- Hermes CLI has cron/scheduling; no official Desktop App yet
- Check which variant each feature belongs to before comparing
關鍵：Memory systems may be read-only
- Codex has a two-phase memory pipeline (rollout extraction + global consolidation)
- But the agent is explicitly instructed: "Never update memories. You can only read them."
- Hermes has bidirectional memory — agent can both read AND write
- This is a meaningful functional difference
關鍵：「Desktop App」嘅定義要查原始碼
- Codex「Desktop」可能只係 npm CLI 包裝，唔係 Electron/Tauri GUI
- Hermes 嘅 Electron app 可能已存在但未公開（
apps/desktop/目錄） - 要 check：有冇
electron/、tauri/、gui/目錄，或者睇 package.json/CMakeLists.txt
陷阱 5：Skills 系統是否自動學習？
- Some agents have static skills (user writes SKILLS.md, never changes)
- Some agents auto-create skills from experience and improve them during use
- Check the source code for auto-learning logic vs static definition
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.
- 8d ago First seen · 121 lines · 7 tokens per session scan A 94472afb0192
agent-capability-comparison-methodology is a skill published in the GitHub repository ChrisLamDev/hermes-core-skills (9 stars, last pushed 2mo ago), licensed MIT. It adds 7 tokens to every session and 1,319 once invoked, about $0.0000 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 skills, from other repositories
unicli-repair
Evidence-driven repair workflow for a broken Uni-CLI adapter. Trigger on a failed unicli envelope, a quarantined adapter, or an explicit adapter-repair request. Classifies non-source failures, edits only the reported adapter path, and uses the original command as a bounded oracle.
autonomous-run
Prepare, start, inspect, resume, or stop a finite local overnight coding run after a human has accepted a Wayfinder terminal spec; coordinates a declared Claude/Codex maker and independent checker without pushing, merging, or writing to external systems.
authentication-patterns
OAuth 2.0, JWT, SSO, MFA, NextAuth/Clerk/Supabase Auth implementation patterns.
case-interview-practice
Interactive consulting case interview practice with structured frameworks, feedback mechanisms, and progressive difficulty. Use when preparing for management consulting interviews, case competitions, or business problem-solving exercises.
i18n-localization
Internationalization and localization for global applications. Use when adding multi-language support, handling regional formats, or preparing apps for global markets.
content-repurposer
Adapt content across platforms with tone/format shifting — blog to social, long to short, text to visual outline. Use when repurposing content for different channels, audiences, or formats.