Tool Call Guard(工具调用安全闸门)
Gate every agent tool call by side effect: reads pass, writes request confirmation, and deletes, external sends and payments are intercepted before they run. 当用户说『agent乱调工具』『误删了文件』『不该发邮件却发了』『怎么给AI的工具加权限边界』,或在给 agent 接工具(文件/网络/数据库/消息/支付)想防危险动作时使用。把每次 tool call 当『需授权操作』:按 读/写/删/外发/支付 分级,危险动作先拦截+提示确认,低
as observed 2026-11-06T15:44:01.797Z- Identifier
tool-call-guard- Source
- ClawHub
- Version observed
- 1.0.12
- Source repository
- not published
- Repository observation
- No source repository listed
- First observed here
- 2026-09-12T14:17:54.398Z
- Observations recorded
- 5
- Installs (reported upstream)
- 1
- Weekly downloads (upstream)
- 622
- Declared license
- MIT-0
Observation history
2026-11-06T15:44:01.797Z
Fields that differed: changelog displayName latestVersion summary
| Field | Before | After |
|---|---|---|
changelog |
"Bilingual rewrite (full EN+ZH, English first)" | "安全面加固:permissions 声明改为与包内脚本实际能力一致;新增「数据处理边界」段,明确默认不外发、需人指定目的地并逐次确认。业务逻辑未改动。" |
displayName |
"tool-call-guard" | "Tool Call Guard(工具调用安全闸门)" |
latestVersion |
"1.0.4" | "1.0.12" |
summary |
"当用户说『agent乱调工具』『误删了文件』『不该发邮件却发了』『怎么给AI的工具加权限边界』,或在给 agent 接工具(文件/网络/数据库/消息/支付)想防危险动作时使用。把每次 tool call 当『需授权操作』:按 读/写/删/外发/支付 分级,危险动作先拦截+提示确认,低风险放行。理论根基:LGD 三律(有籍·有证·有门禁)。触发词:工具调用安 | "Gate every agent tool call by side effect: reads pass, writes request confirmation, and deletes, external sends and payments are intercepted before they run. 当用户说『agent乱调工具』『误删了文件 |
2026-10-17T20:48:50.018Z
Fields that differed: changelog latestVersion
| Field | Before | After |
|---|---|---|
changelog |
"权利层第 3 轮:① §3.2 权属宣告统一块归一——清除块内他资产事实(此前误填 uibc-core 的 DOI / commit cb6f11b / Sigstore Rekor 锚),改按本资产真值填写;② 技能件许可口径过授权清除——理论文本由 CC BY 4.0 改为「保留所有权利」,.py 依 §3.2 标 MIT;③ 保留分层许可:代码 MI | "Bilingual rewrite (full EN+ZH, English first)" |
latestVersion |
"1.0.3" | "1.0.4" |
2026-09-29T04:47:03.526Z
Fields that differed: changelog latestVersion
| Field | Before | After |
|---|---|---|
changelog |
"- Added version 1.1.0 with updated documentation in SKILL.md.\n- Clarified use cases: intercepts and confirms agent tool calls by risk level (read/write/delete/forward/payment).\n | "权利层第 3 轮:① §3.2 权属宣告统一块归一——清除块内他资产事实(此前误填 uibc-core 的 DOI / commit cb6f11b / Sigstore Rekor 锚),改按本资产真值填写;② 技能件许可口径过授权清除——理论文本由 CC BY 4.0 改为「保留所有权利」,.py 依 §3.2 标 MIT;③ 保留分层许可:代码 MI |
latestVersion |
"1.0.0" | "1.0.3" |
2026-09-17T19:21:47.070Z
Fields that differed: license description
| Field | Before | After |
|---|---|---|
license |
null | "MIT-0" |
description |
"当用户说『agent乱调工具』『误删了文件』『不该发邮件却发了』『怎么给AI的工具加权限边界』,或在给 agent 接工具(文件/网络/数据库/消息/支付)想防危险动作时使用。把每次 tool call 当『需授权操作』:按 读/写/删/外发/支付 分级,危险动作先拦截+提示确认,低风险放行。理论根基:LGD 三律(有籍·有证·有门禁)。触发词:工具调用安 | null |
Correction
If you maintain this extension and believe anything above is inaccurate, request a correction. Corrections are published, and disputed entries are marked as disputed while under review.