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 ga626/knowledge-radar --skill researchgit clone --depth 1 https://github.com/ga626/knowledge-radarWrote 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/ga626/knowledge-radar/research)<a href="https://agentmods.dev/skills/ga626/knowledge-radar/research"><img src="https://agentmods.dev/badge/skills/ga626/knowledge-radar/research/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/ga626/knowledge-radar/research"><img src="https://agentmods.dev/badge/skills/ga626/knowledge-radar/research.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 | $0.00033 | $0.00837 |
| Opus 5.5 | $0.00013 | $0.00335 |
| Sonnet 5.5 | $0.00007 | $0.00167 |
| Haiku 4.5 | $0.00003 | $0.00084 |
Grade A, and why
research 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 1mo 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
KnowledgeRadar 调研
KnowledgeRadar 是 Codex 的外部研究感知层；用户级 config.toml 是唯一 MCP 注册源，本插件只提供工作流，不注册第二个 server。
工作流
- 优先使用原生
mcp__knowledgeradar.*。未显示时，先发现knowledgeradar mcp health_check kr_research get_capabilities；发现仍失败就记录“本任务原生工具面未附着”，而不是把服务健康误报为可用。原生明确Transport closed后，先用项目协议探针区分后端与当前 Desktop MCP session，不重启共享 KR 服务。若宿主暴露官方config/mcpServer/reload，先以mark-host-refresh-pending记录请求，再在下一 user turn 只用真实原生health_check(summary)与get_capabilities(summary=true)两项共同验收；缺一项均不是恢复。若宿主未暴露刷新入口，完整退出并重新启动 Codex Desktop 是可重复的宿主恢复动作，但同样必须完成原生双调用验收。任务需要继续时，直接使用版本化scripts\kr_mcp_continuity.py call；它绑定当前active.json安装身份，对只读就绪调用最多新建一次重试会话，并输出access_path=continuity_fallback、artifact identity、工具目录 fingerprint 与 retry receipt。不得宣称原生 MCP 已恢复，也不得自动重放写入、账号或浏览器操作。 - 非平凡或不确定任务先调用
health_check(mode="summary")与get_capabilities(summary=true)，再由 Agent 自主选择来源生态、轮次、工具、扩展和停止点。 - 重度研究优先用
kr_research，后续按证据需要调用已注册的 Web、学术、GitHub、视频、平台、详情、任务状态和决策日志工具。 - 内置 web/search 只能在 KR 路由后作为
host_internal_web_wave使用，并记录wave_id、strategy_tree、reason和relationship_to_kr。 - 严肃报告记录证据登记：访问路径、工具、来源 URL/路径、事实、推理、强弱和缺口。搜索结果只是候选，需详情抽取或交叉来源确认后才支持强结论。
- 一份报告对应一个
research_task，完成时使用同一任务 ID 调用finalize_research_task；每个考虑过的来源生态都记录使用、跳过、阻断、不相关或未到达及原因。 - 深度或治理报告须通过项目研究质量检查并完成最终化；sidecar 是声明，不等于每次工具调用的证明，关联不清时标为 partial。
失败边界
- 原生工具在发现后仍不可用：带
--reason、任务 ID、回执和 handoff 的版本化continuity_fallback是正式连续性路径；它只保证任务连续，不能让当前线程重新获得原生工具。用户完整重启 Desktop 后，只有在同一 artifact 上重新发生health_check与get_capabilities两次真实 native call，才能写native_mcp_restored_after_desktop_restart。不调用未版本化、身份不明或工具目录不完整的 fallback。 - 缺少工具卡不等于 KnowledgeRadar 服务故障；区分宿主注入、stdio、HTTP 和源码配置。
- 登录、验证码和平台验证转为明确的人工交互状态，不盲目循环重试。
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.
- 1mo ago Changed 5c69c032bb99
- 1mo ago First seen · 25 lines · 33 tokens per session scan A 85aa33e69504
research is a skill published in the GitHub repository ga626/knowledge-radar (0 stars, last pushed 11d ago), licensed MIT. It adds 33 tokens to every session and 837 once invoked, about $0.0001 per session on Opus 5.5. 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
status-card-query
Create and maintain agent-authored Paperclip status cards, or compile a prose interest prompt into bounded CompanySearchQuery objects and write the first summary from the assigned Summarizer run.
gpt-researcher
Autonomous deep research from Codex via MCP.
bilibili_search
Search for videos on Bilibili using keyword queries generated from user interests.
vs-search-tuning-partial-case
Use when the user provides 1-50 concrete bad-case search queries for one Viking Search app and wants local deterministic fixes. This skill only verifies request-level fine-operation interventions against a read-only baseline scene and delivers a console-ready configuration sheet, validated payloads, and a replay…
site-config-creation
Create or update the [site] section of stencila.toml for published Stencila sites. Use when asked to configure site domain, title, author, logo, icons, labels, descriptions, socials, featured content, navigation, routes, access, layout presets and regions, glide, search, formats, reviews, uploads, remotes, actions…
vs-search-tuning
Use when a user asks an agent to evaluate or tune text search similarity for an existing Viking AI Search application and dataset.