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 aAAaqwq/AGI-Super-Team --skill binance-squaregit clone --depth 1 https://github.com/aAAaqwq/AGI-Super-TeamWrote 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/aaaaqwq/agi-super-team/binance-square)<a href="https://agentmods.dev/skills/aaaaqwq/agi-super-team/binance-square"><img src="https://agentmods.dev/badge/skills/aaaaqwq/agi-super-team/binance-square/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/aaaaqwq/agi-super-team/binance-square"><img src="https://agentmods.dev/badge/skills/aaaaqwq/agi-super-team/binance-square.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00102 | $0.06210 |
| Opus 5 | $0.00051 | $0.03105 |
| Sonnet 5 | $0.00020 | $0.01242 |
| Haiku 4.5 | $0.00010 | $0.00621 |
Grade A, and why
binance-square 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.
How it starts
The opening of the file, as written. The whole thing — 273 lines — stays where its author put it; the contents beside it link to each section on GitHub.
币安广场合约投机雷达 v5.0(Shadow)
当前可真实运行的是只读影子报告:不发 Telegram、不创建定时任务、不下单。
必读合同
执行前完整读取:
GRILL_DECISIONS.md:用户确认的产品、作者、行情、评分、追踪与留存规则。V4_IMPLEMENTATION_PLAN.md:里程碑、证据门和已知阻塞。
不得把设计目标描述成已经上线的能力。最终结论必须来自当次生成的 JSON/Markdown 和测试结果。
当前已实现
-
v5.0 对候选分层做了探索性放宽:
TOP >= 55、WATCH >= 40;缺少 Entry/SL/TP 时按字段完整度部分计分,不再仅因缺字段进入评分硬过滤,但执行就绪状态保持UNAVAILABLE,不能成为可执行 TOP。TP1 RR1.2与主要目标 RR1.5是评分线;主要目标低于1.5仍硬过滤。 -
v5.0 增加
--scheduled-retry入口和最多两次相互独立的 Feed 捕获;重试不得跨次拼接帖子或复用旧快照冒充本轮成功。 -
Telegram renderer 改为面向用户的简洁摘要;完整 JSON/Markdown、manifest 与 SQLite 仍是唯一审计事实源,精简消息不能替代证据文件。
-
标准入口默认容量 200;Profile 是专业作者帖主通道,Feed 与经过人工/模型提取的参数信号 URL 作有界补充,按 canonical
post_id去重且保留全部来源 observation。 -
Feed 使用
square-feed-coverage/v1:必须从顶部开始,并记录滚动几何、两相懒加载触发、停止原因和覆盖状态。当前 scope 固定为BINANCE_SQUARE_DISCOVER_DOM,global_denominator_known=false且pagination_api_exhaustion_verified=false;即使达到声明下限并稳定穷尽 DOM 而标BOUNDED_COMPLETE,也只代表该次 DOM surface,不代表内部 feed API 或整个平台。容量、滚动预算、错误或未验证旧快照分别标CAPPED/PARTIAL/BLOCKED/LEGACY_UNVERIFIED。 -
Feed 每次成功观察都写不可变快照和
binance_raw_posts.jsonobserved latest。capture_attempt_no、capture_attempt_limit与表示稳定物理底部事实的surface_exhausted是square-feed-coverage/v1的向后兼容可选诊断扩展,旧 v1 证据不会被静默升级。只有严格满足BOUNDED_COMPLETE + EXHAUSTED + minimum_target_met + unique>=preferred_minimum的 Discover DOM 捕获才推进独立binance_raw_posts_eligible.json。PARTIAL 证据必须保存但不得覆盖既有 eligible 指针;抓取锁忙以非零状态失败,不能复用旧 latest 冒充本轮成功。 -
Feed 文本在 CDP 入站时执行 Unicode scalar 规范化:合法 emoji 代理对保持原义,孤立 UTF-16 surrogate 仅替换为
U+FFFD并计入malformed_unicode_replacements。最终 JSON 写入层再次递归兜底;写入失败必须清除临时文件,单个异常字符不得拖垮整批帖子。 -
生产 wrapper 默认最多执行两次相互独立的 Feed 捕获(
--capture-attempt-limit 1可回滚为单次),首次 eligible 立即停止;两次都未 eligible 时只选择最新一次合法捕获进入 radar,不跨次拼接帖子,也不读取旧 eligible。PARTIAL与达到 200 条硬上限的CAPPED都只作带原始覆盖标签的不完整 Feed 补充,CAPPED不得冒充穷尽。每次 observed latest 都必须刷新、与当次 immutable snapshot 逐字节一致并通过 v1 合同;只有 eligible 捕获才额外要求 eligible pointer 逐字节一致。Feed 是 Profile 主通道的补充,合法不完整捕获不再阻断 Profile,但下游仍必须如实保留覆盖状态;malformed、未刷新、错误或指针不一致继续失败关闭。 -
逐帖请求币安公开详情接口,使用稳定
post_id、author_id、权威正文与精确发布时间。 -
Profile + Feed 双通道已接入真实只读 pipeline:Profile 计划只认稳定
squareUid,通过匿名公共 GET 从timeOffset=-1分页;逐条用firstReleaseTime执行严格24小时窗口,并以可验证 cursor 水位、空页或无 next 作为终止证明。置顶旧帖不会造成提前停止,cursor 停滞/上升、锚点不匹配、schema 漂移、请求失败或页预算耗尽均失败关闭为PARTIAL。 -
历史 fixture 保留 30D
PNL/ROI/VOLUME/WIN_RATE四榜解析合同;当前真实 Smart Money 接入只支持官方 Web UI 已观测的PNL/ROI,每榜必须恰好30个唯一排名与topTraderId才可标记COMPLETE。 -
Smart Money 公共只读 API 已接入标准 pipeline:
--smart-money在线采集,--smart-money-fixture注入带完整性校验的证据;排行、交易员详情与 Square 身份映射分别计数,不能相互代替。 -
单次生产影子入口
scripts/run_production_cycle.py会按上述独立捕获规则刷新 Feed,再以前台、顺序、--no-send方式运行 Smart Money、官方新闻与雷达;Feed 快照超过10分钟或来自未来时失败关闭。离线测试只验证选择与失败关闭合同,尚无真实双捕获稳定性回执。 -
生产 wrapper 为每个长时间子进程每30秒输出一次可刷新的
[HEARTBEAT],避免调度器把正常静默分析误判为无输出挂死;父进程异常退出时会终止并回收子进程。run_radar.py将SIGTERM转成可审计异常,使已创建的 attempt 进入FAILED,而不是永久遗留RUNNING。 -
scripts/render_latest_telegram.py --report <committed-report.json>只读取已提交报告并复用标准render_telegram合同输出精简文本;它本身不发送。账号、目标、调度与实际发送仍属于安装环境的外部配置,不随公共 Skill 发布。 -
生产入口默认不再消费旧版可变
data/signal_check_input.json;只有显式传入--signals-json才把它作为有manifest的补充来源。真实 pipeline 在任何 Profile、Smart Money 或帖子详情网络请求之前获取一次 Futures catalog 并完成 Binance server-time 校验,后续行情复用同一 catalog。 -
官方 Binance 公告可用
--official-news纳入严格前24小时来源覆盖;列表缺发布时间时必须回源详情,响应 envelope、文章 code、新到旧排序和窗口穷尽性均受合同约束,限流或解析失败会留下FAILED来源状态而不是伪造空新闻。 -
Smart Money fixture 导入不只校验可重签的 payload hash,还会重新验证固定语义:
30D、DESC、恰好PNL+ROI、每页10行×3页,以及两个onlyShow*filter 均为false;任一不符即拒绝。 -
topTraderId → squareUid身份映射 v2 必须来自同一匿名 Binance 响应,并绑定 raw bytes、request manifest、实际 SHA-256 和/data/topTraderId、/data/squareUidJSON Pointer。采集器只能生成PROPOSED;只有经独立、显式 review 生成的不可变APPROVEDartifact 才能进入活动投影,且支持事件化撤销。 -
显式映射证据或经哈希固定的多映射 catalog 可通过
--smart-money-square-mapping-evidence接入同一 pipeline;报告把 Smart Money 映射/Profile 与独立 seed Profile 分开记分母,seed 覆盖不得冒充60人榜单覆盖。 -
严格最近 24 小时窗口:
[decision_at - 24h, decision_at)。 -
旧 signal 文件只提供作者原始 Entry/SL/TP 等参数;时间与作者必须回源重验,标记为
AUTHOR_LEGACY_EXTRACTED。 -
DERIVATION_POLICY_V1:作者参数有效时保留AUTHOR;缺参数或入场已错过时,只有完整 Futures 四周期证据、结构、0.4 ATR 止损、RR 1.5/2.0 与禁追价门全部通过,才生成显著标记的SYSTEM_REDERIVED;否则REJECTED。 -
作者原始计划会从发布时间回放到决策时点;期间先触发 SL/TP 即拒绝,同根 K 线同时覆盖 SL/TP 时保守按止损优先。作者止损还必须通过发布前结构和最小 0.4 ATR 距离门。
-
布林带收缩但尚未放量突破时降为 WATCH;只有方向一致且量能确认的突破才允许继续参与 TOP 判定。
-
共识按 symbol/direction、稳定作者ID与内容哈希建立;候选自身、同作者多帖和复制内容簇不增加共识分。
-
未验证作者不冒充 Tier A,作者可信度为 0 分。
-
精确验证
TRADING / PERPETUAL / USDT合约。 -
每次真实运行保存完整 Futures 合约目录证据,不只保存本轮涉及的币种。
-
Futures 公共行情:现价、标记价、指数价、资金费率、持仓量、24h量、15m/1h/4h/1d 已收盘 K 线。
-
Futures 暂时不可用时,只允许 Spot Proxy 补价格和 OHLC;不得用 Spot 量冒充合约量。
-
指标:BB20、%B、BandWidth、ATR14、Futures volume ratio、多周期趋势。
-
固定 100 分评分、硬门槛、TOP3/WATCH5、同币多空冲突降级。
-
手续费、滑点、资金费率和深度模型默认未配置;只有完整注入成本参数且成本调整后主要目标 RR
>=2才可标记执行就绪,不能把名义 RR 报成净 RR。 -
本地不可变 raw/market/report JSON、Markdown、单一
latest目录指针原子替换、SQLite logical run/attempt/manifest。 -
生产 logical run 仍由固定 UTC 四小时 slot 唯一标识;首次 attempt 的
decision_at取 attempt-start UTC,并必须由 Binance server time 在30秒偏差内验证、且位于 slot 后 0–10 分钟。验证失败则market_catalog/FAILED,禁止用本机时钟降级成功。SQLiteproduction_run_cutoff每轮只写一次;唯一一次 +10 分钟 retry 必须精确复用,不能改写 cutoff。帖子资格与已闭合 K 线以它为 cutoff;标量行情是随后抓取的实时验证值,必须单独显示captured_at。 -
job_namespace + production_job_id + scheduled_for_utc共同隔离 production/canary 身份,避免不同账本生成相同逻辑运行或 attempt 标识。 -
logical run 与
RUNNINGattempt 在任何输入文件、catalog 或帖子详情采集之前落库;FETCH/分析/报告失败均终结为带阶段的FAILED,retry只能从一个已审计的 FAILED attempt1 分配 attempt2。 -
生产跨轮去重只读取更早且
SUCCEEDED的同 job 时槽;FAILED/RUNNING 不污染基线。回放默认明确显示NOT_COMPUTED。 -
v4.1 增量数据合同与追踪持久化继续保留:signal family/revision、PENDING/触发/结果、MFE/MAE/R;当前轮失败后查询自动排除其状态。
-
UTC 六时槽、一次 +10 分钟重试、静默恢复与二次失败分类告警已有纯 shadow runner;重试成功仍生成该轮唯一的常规
HELD/NO_SEND报告 intent,但不生成恢复告警;不安装任务、不发送。 -
报告分别记录广场抓取开始/完成、行情最新和报告生成时间;K线请求冻结
endTime=decision_time_ms-1,跨周期边界仍保持点时一致。 -
报告固定对账 Feed、Profile、Smart Money、官方新闻、行情目录五类来源;每类都显示
COMPLETE/PARTIAL/FAILED/NOT_ATTEMPTED、provenance、采集时间和证据清单,不能用一个来源替代另一个来源的完成度。 -
Profile报告必须同时携带Smart Money与seed cohort分母;planned、五类终态、逐作者outcome、稳定身份、顶层状态与source coverage必须对账且唯一。覆盖率分子仅为
COMPLETE+EMPTY,任何PARTIAL都不算完成;全EMPTY表示已穷尽无帖,归一为COMPLETE。
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.
- AUTHORIZATION_RECEIPT.md 2.1 KB
- GRILL_DECISIONS.md 29 KB
- migrations/001_v4.sql 3.7 KB
- migrations/002_contract_freeze.sql 9.0 KB
- migrations/003_production_cutoff.sql 818 B
- migrations/004_smart_money_leaderboards.sql 9.7 KB
- migrations/005_run_namespace_and_lineage.sql 957 B
- migrations/006_identity_mapping_v2.sql 3.6 KB
- migrations/007_profile_fetch_outcomes.sql 2.3 KB
- migrations/008_post_discovery_observations.sql 1.7 KB
- OPTIMIZATION_VERIFICATION.md 11 KB
- PRE_OPTIMIZATION_AUDIT.md 1.7 KB
- PROVENANCE.md 1.0 KB
- scanner/__init__.py 134 B runs code
- scanner/authors.py 48 KB runs code
- scanner/binance_news.py 7.2 KB runs code
- scanner/contracts.py 12 KB runs code
- scanner/derivation.py 21 KB runs code
- scanner/discovery.py 28 KB runs code
- scanner/indicators.py 6.3 KB runs code
- scanner/market.py 21 KB runs code
- scanner/opportunities.py 15 KB runs code
- scanner/pipeline.py 118 KB runs code
- scanner/profile_pipeline.py 19 KB runs code
- scanner/reports.py 62 KB runs code
- scanner/retention.py 2.1 KB runs code
- scanner/runner.py 20 KB runs code
- scanner/scoring.py 27 KB runs code
- scanner/smart_money.py 20 KB runs code
- scanner/square.py 16 KB runs code
- scanner/storage.py 150 KB runs code
- scanner/tracking.py 17 KB runs code
- scripts/binance_price_check.py 9.1 KB runs code
- scripts/binance_scraper.py 39 KB runs code
- scripts/binance_state_commit.py 3.0 KB runs code
- scripts/build_identity_mapping_catalog.py 6.4 KB runs code
- scripts/cleanup_retention.py 3.3 KB runs code
- scripts/collect_identity_mapping.py 9.6 KB runs code
- scripts/collect_smart_money.py 3.1 KB runs code
- scripts/collect_square_v4.py 25 KB runs code
- scripts/discover_authors.py 5.7 KB runs code
- scripts/render_clean.py 4.2 KB runs code
- scripts/render_latest_telegram.py 1.6 KB runs code
- scripts/repair_latest.py 1.5 KB runs code
- scripts/review_identity_mapping.py 8.7 KB runs code
- scripts/run_production_cycle.py 13 KB runs code
- scripts/run_radar.py 7.8 KB runs code
- scripts/run_shadow_cycle.py 4.6 KB runs code
- scripts/scan_swing_opportunities.py 13 KB runs code
- tests/__init__.py 52 B runs code
- tests/fixtures/discovery/leaderboard_seed_profiles.json 3.5 KB
- tests/fixtures/discovery/leaderboards_empty.json 25 B
- tests/fixtures/discovery/profile_contents_public.json 650 B
- tests/fixtures/history/v3_snapshot_01.json 413 B
- tests/fixtures/history/v3_snapshot_02.json 560 B
- tests/fixtures/history/v3_snapshot_03.json 710 B
- tests/fixtures/m1b/feed_cards_live.json 1.9 KB
- tests/fixtures/m1b/post_detail_live.html 1.5 KB
- tests/fixtures/m1b/post_detail_public.json 1.2 KB
- tests/fixtures/m1b/profile_live.html 1.0 KB
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 · 273 lines · 102 tokens per session scan A fbc2f14168b3
binance-square is a skill published in the GitHub repository aAAaqwq/AGI-Super-Team (91 stars, last pushed 2d ago), licensed MIT. It adds 102 tokens to every session and 6,210 once invoked, about $0.0005 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 2026-09-05.
Other skills, from other repositories
blocknative-openapi-skill
Operate Blocknative gas intelligence APIs through UXC with a curated OpenAPI schema, API-key auth, and read-first guardrails.
coingecko-openapi-skill
Operate CoinGecko and GeckoTerminal market data APIs through UXC with a curated OpenAPI schema, API-key auth, and read-first guardrails.
alchemy-openapi-skill
Operate Alchemy Prices API reads through UXC with a curated OpenAPI schema, path-templated API-key auth, and read-first guardrails.
defillama-pro-openapi-skill
Operate DefiLlama Pro analytics APIs through UXC with a curated OpenAPI schema, path-templated API-key auth, and read-first guardrails.
helius-openapi-skill
Operate Helius Wallet API reads through UXC with a curated OpenAPI schema, API-key auth, and read-first guardrails.
mexc-openapi-skill
Operate MEXC Spot REST APIs through UXC with a curated OpenAPI schema, HMAC query signing, and separate public/signed workflow guardrails.