Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Tencent-RTC/agent-skillsnpx agentmods add skills/tencent-rtc/agent-skills/trtc-ai-realtime-interpreterWrote 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/tencent-rtc/agent-skills/trtc-ai-realtime-interpreter)<a href="https://agentmods.dev/skills/tencent-rtc/agent-skills/trtc-ai-realtime-interpreter"><img src="https://agentmods.dev/badge/skills/tencent-rtc/agent-skills/trtc-ai-realtime-interpreter/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tencent-rtc/agent-skills/trtc-ai-realtime-interpreter"><img src="https://agentmods.dev/badge/skills/tencent-rtc/agent-skills/trtc-ai-realtime-interpreter.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00119 | $0.11622 |
| Opus 5 | $0.00060 | $0.05811 |
| Sonnet 5 | $0.00024 | $0.02324 |
| Haiku 4.5 | $0.00012 | $0.01162 |
Grade B, and why
trtc-ai-realtime-interpreter scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
| 写完 .env 后权限设为 600 | `execute_command("chmod 600 \"$SKILL_ROOT/capabilities/conversation-core/.env\"")` | Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
sleep 8 && curl -k -sS https://localhost:8020/api/v1/health How it starts
The opening of the file, as written. The whole thing — 728 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TRTC AI Realtime Interpreter Skill (v1.0)
本文档是 Coding Agent 的执行 SOP，也是面向用户的参考指南。 任何涉及"实时翻译 / AI 会议翻译 / 用 TRTC 做同声传译"的自然语言意图，都应先读本文件再动手。 所有脚本调用必须严格遵守 §11 工具白名单。
0. 路径基线（SKILL_ROOT / PROJECT_ROOT）—— 最高优先级，先读这里
本 Skill 的所有运行时资产（capabilities/、scripts/、scenarios/、auto_adapters/、start.sh）
都在 Skill 自己的目录下，不一定在用户工作区根目录。Skill 可能被安装在任意位置：项目子目录、
.agents/skills/、.codebuddy/skills/ 等。因此永远不要假设"Skill 根目录 == 工作区根目录"。
| 变量 | 含义 | 如何获取 |
|---|---|---|
SKILL_ROOT |
Skill 自己的目录（包含 SKILL.md / scripts/ / capabilities/ ...） |
= 加载本 Skill 时系统注入的 Base directory 绝对路径。Agent 需要记住它。 |
PROJECT_ROOT |
用户当前项目根目录（= 工作区根目录；Path B 的集成目标） | = 当前工作区根目录的绝对路径。 |
硬性规则：
- 所有调用 Skill 内置脚本/资产的命令必须用
SKILL_ROOT的绝对路径，例如：
不要写裸相对路径（假设它相对工作区根目录解析）——这是历史版本最常见的 bug 根源。cd "$SKILL_ROOT" && python3 scripts/verify-credentials.py - 本文档所有含
$SKILL_ROOT/$PROJECT_ROOT的命令模板，Agent 执行前必须替换成真实绝对路径。 - 如果一时确定不了
SKILL_ROOT，做一次性探测（不要让用户挪目录）：find "$PWD" -maxdepth 4 -name SKILL.md -path '*trtc-ai-realtime-interpreter*' 2>/dev/null | head -1
1. 何时使用本 Skill
触发条件（见文件头 frontmatter 的 triggers.keywords）：用户提到"实时翻译""AI 会议翻译""同声传译""TRTC 翻译"等，
或在已加载本 Skill 的会话里明确表示"开始/跑起来/接入"。
不适用（应拒绝并说明）：
- 纯文字翻译（不涉及语音/实时通话）
- 离线文档翻译
- 非 TRTC 生态的实时音视频方案（应引导至对应 Skill）
产品定位说明：本 Skill 封装的是 TRTC Conversational AI（语音）实时翻译能力，默认演示场景是"多人会议室， 谁说话都能被实时翻译"。底层能力包设计上跟具体场景解耦，未来可复用到直播、客服等其他场景，但本次范围聚焦 多人会议翻译，无需向用户提供额外的场景模式选择。
2. 交互语言检测（全程硬性约束）
跟随触发本 Skill 的第一条用户消息判断语言（中文/英文/其他近似英文），存入 interaction_lang；
用户中途明确要求切换语言则立即更新。所有引导文案、ask_followup_question 的问题与选项、
错误提示，均需遵循 interaction_lang。
Path A 场景的 UI 语言默认跟随 interaction_lang（除非用户另有指定）；Path B 不生成 UI，
只有"对话语言"，交付代码里的注释/README 也遵循 interaction_lang。
2.5 Prompt Reporting
- Root/host bootstrap 统一处理 Prompt、路由和 Host Stop notice；若本 Skill 被直接调用且 Host 未记录本轮，才用 stdin Prompt 入口补一次。普通失败继续业务流程；固定控制 marker 按 root dispatcher 规则处理。
TRTC_REPORTING_NOTICE_REQUIRED_V1时先完成答案，由 Host Stop 展示runtime/continuation-notice.md，模型不得自行附加或改写。 - 展示澄清问题前运行同脚本的
context --question "<完整问题>"；固定选项仍使用标准选项 UI。
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- .gitignore 398 B
- auto_adapters/integration_templates/generic-rest-api.md 3.7 KB
- auto_adapters/integration_templates/room-owner-authz-note.md 2.6 KB
- auto_adapters/manifest.yaml 2.8 KB
- auto_adapters/python/fastapi_reverse_proxy.py.tpl 3.4 KB
- auto_adapters/python/README.md 710 B
- capabilities/conversation-core/.env.example 449 B
- capabilities/conversation-core/frontend/silent-listener.ts 2.9 KB runs code
- capabilities/conversation-core/manifest.yaml 4.0 KB
- capabilities/conversation-core/requirements.txt 107 B
- capabilities/conversation-core/src/__init__.py 0 B runs code
- capabilities/conversation-core/src/_capability_loader.py 3.1 KB runs code
- capabilities/conversation-core/src/agent.py 5.8 KB runs code
- capabilities/conversation-core/src/credentials.py 3.3 KB runs code
- capabilities/conversation-core/src/health.py 10 KB runs code
- capabilities/conversation-core/src/log_filter.py 1.0 KB runs code
- capabilities/conversation-core/src/server.py 10 KB runs code
- capabilities/conversation-core/src/trtc_client.py 8.0 KB runs code
- capabilities/conversation-core/src/usersig.py 1.8 KB runs code
- capabilities/meeting-ops/manifest.yaml 3.4 KB
- capabilities/meeting-ops/README.md 2.2 KB
- capabilities/meeting-ops/src/__init__.py 0 B runs code
- capabilities/meeting-ops/src/fanout.py 6.6 KB runs code
- capabilities/meeting-ops/src/router.py 3.0 KB runs code
- capabilities/realtime-translation/frontend/subtitle-parser.ts 4.6 KB runs code
- capabilities/realtime-translation/manifest.yaml 3.2 KB
- capabilities/realtime-translation/src/__init__.py 0 B runs code
- capabilities/realtime-translation/src/modes.py 3.1 KB runs code
- capabilities/realtime-translation/src/router.py 2.7 KB runs code
- capabilities/realtime-translation/src/service.py 3.1 KB runs code
- README.ja.md 15 KB
- README.md 12 KB
- README.zh-CN.md 12 KB
- scenarios/meeting-interpreter/backend/app/__init__.py 0 B runs code
- scenarios/meeting-interpreter/backend/app/server.py 12 KB runs code
- scenarios/meeting-interpreter/backend/requirements.txt 104 B
- scenarios/meeting-interpreter/backend/start.sh 1.0 KB runs code
- scenarios/meeting-interpreter/README.md 1.7 KB
- scenarios/meeting-interpreter/recipe.yaml 3.1 KB
- scenarios/meeting-interpreter/ui/.gitignore 88 B
- scenarios/meeting-interpreter/ui/index.html 397 B
- scenarios/meeting-interpreter/ui/legacy/index.html 40 KB
- scenarios/meeting-interpreter/ui/package-lock.json 154 KB
- scenarios/meeting-interpreter/ui/package.json 921 B
- scenarios/meeting-interpreter/ui/postcss.config.js 80 B runs code
- scenarios/meeting-interpreter/ui/src/api/backend.ts 2.5 KB runs code
- scenarios/meeting-interpreter/ui/src/App.vue 568 B
- scenarios/meeting-interpreter/ui/src/components/conference/ChatPanel.vue 5.0 KB
- scenarios/meeting-interpreter/ui/src/components/conference/ConferenceRoom.vue 12 KB
- scenarios/meeting-interpreter/ui/src/components/conference/ParticipantViewUI.vue 5.0 KB
- scenarios/meeting-interpreter/ui/src/components/conference/PeoplePanel.vue 6.2 KB
- scenarios/meeting-interpreter/ui/src/components/conference/SidePanel.vue 1.9 KB
- scenarios/meeting-interpreter/ui/src/components/conference/Toolbar.vue 10 KB
- scenarios/meeting-interpreter/ui/src/components/conference/TopBar.vue 8.0 KB
- scenarios/meeting-interpreter/ui/src/components/conference/TranscriptPanel.vue 5.9 KB
- scenarios/meeting-interpreter/ui/src/components/SetupScreen.vue 10.0 KB
- scenarios/meeting-interpreter/ui/src/components/SummaryScreen.vue 5.6 KB
- scenarios/meeting-interpreter/ui/src/composables/useAiInterpreter.ts 7.9 KB runs code
- scenarios/meeting-interpreter/ui/src/composables/useConference.ts 3.1 KB runs code
- scenarios/meeting-interpreter/ui/src/env.d.ts 186 B runs code
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.
- 6d ago Changed 13bef963927a
- 11d ago First seen · 728 lines · 119 tokens per session scan B 2818966850e5
trtc-ai-realtime-interpreter is a skill published in the GitHub repository Tencent-RTC/agent-skills (13 stars, last pushed 2d ago), licensed MIT. It adds 119 tokens to every session and 11,622 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
cometchat-native-theming
CometChatThemeProvider + CometChatI18nProvider — color tokens, typography, dark mode, per-component style overrides, and localization (18 built-in languages + custom translations). The JS theme object replaces CSS variables.
cometchat-i18n
Localization (i18n) across all CometChat UI Kit families — React, React Native, Angular, Android (V5/V6), iOS, Flutter (V5/V6). Covers CometChatLocalize.init signature differences (positional vs object), bundled languages, custom-language registration, RTL support, fallback to English, and cross-family drift risks.…
deepinit
../../../.agents/workflows/deepinit.md.
chinese-documentation
A Chinese technical-documentation style guide covering spacing, punctuation, numbers, terminology, and links when Chinese and English appear together.
seedance-vocab-ja
This skill should be used when the user asks for Japanese Seedance 2.0 prompt wording, Japanese cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms into Japanese.
tiktok-shop-cross-border
Cross-border selling on TikTok Shop. Market selection, logistics setup, localization strategy, compliance requirements, and international expansion. Use when the user asks about TikTok Shop international selling, cross-border ecommerce, global expansion, or selling in multiple countries.