Skills/ClawHub/anzhuang-suanliang

anzhuang-suanliang@0.4.39

CAUTION· install only

Publisher claims: "安装出概算与物资计划;不出金额、不替代结算。触发词:安装算量、桥架、喷淋。" Observed in sandbox: sends data to release-assets.githubusercontent.com.

Registry ClawHubPublisher lisong2003-lgtm · publishing for 38 daysInstalls 962Last detonated 28 min ago

Coverage

Bundled scripts ran with no arguments; no agent task scenario for this bundle.
Tested
install scripts, with decoy credentials in place · second launch with the clock 45 days ahead · 8 of 19 bundled scripts run with no arguments (the rest were not run: per-bundle cap), with decoy credentials in the environment · every file read, file written and host contacted by what ran
Not tested
use of the skill by an agent

What changed since 0.4.35

vet-skills diff anzhuang-suanliang 0.4.35 0.4.39

This version reaches for something 0.4.35 did not: 2 new hosts (github.com, release-assets.githubusercontent.com) · 1 new install script (bash ./scripts/setup.sh). Compared from the two recorded runs, not from source.

New hosts
github.com, release-assets.githubusercontent.com
New install scripts
bash ./scripts/setup.sh

0.4.35 was not run in the sandbox (static only); the comparison covers what static analysis recorded only.

The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (3 scenarios; bundled scripts ran with no arguments; no agent task scenario for this bundle) on 2026-11-18 and scored it 56/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Privileged sudo command. Summary: sends data to release-assets.githubusercontent.com.

Behavior trace

Honeypot run · 1 s · 3 scenarios
t+CategoryObserved action
0.0sprocessinstall script run in the sandbox
0.4sfilesystemread ~/.workbuddy/skills/ outside project
0.4snetworkCONNECT tcp://github.com:443
0.4snetworkCONNECT tcp://release-assets.githubusercontent.com:443 · not in the allow-list, refused
0.4sprocessexec /usr/bin/dirname
0.4sprocessexec /usr/bin/find
0.4sprocessexec /app/apps/web/node_modules/.bin/chmod
0.4sprocessexec /root/.cache/node/corepack/v1/pnpm/10.28.0/dist/node-gyp-bin/chmod
0.4sprocessexec /app/node_modules/.bin/chmod
0.4sprocessexec /pnpm/chmod
0.4sprocessexec /usr/local/sbin/chmod
0.4sprocessexec /usr/local/bin/chmod
0.4sprocessexec /usr/sbin/chmod
0.4sprocessexec /usr/bin/chmod
0.4sprocessexec /usr/bin/mktemp
0.4sprocessexec /usr/bin/curl
0.4sprocessexec /usr/bin/rm
0.4sprocessexited 3: [安装算量] 下载底座失败。离线请设 SKILL_NO_BASE=1 降级使用;联网重试建议 export CAD_FILE_READER_VERSION=0.34.1 后重试。
0.4sprocessbundled script run in the sandbox with no arguments
0.4sprocessexited 2: [anzhuang statesync] 找不到共享模块 cad_statesync_core.py(试过 CAD_FILE_READER_SCRIPTS / 融合包 modules / 技能根)
0.4sprocessbundled script run in the sandbox with no arguments
0.5sfilesystemread ~/.cache/codex-runtimes/ · ~/.workbuddy/skills/ outside project
0.5sprocessexec /usr/bin/dirname
0.5sprocessexec /usr/bin/find
0.5sprocessexec /app/apps/web/node_modules/.bin/chmod
0.5sprocessexec /root/.cache/node/corepack/v1/pnpm/10.28.0/dist/node-gyp-bin/chmod
0.5sprocessexec /app/node_modules/.bin/chmod
0.5sprocessexec /pnpm/chmod
0.5sprocessexec /usr/local/sbin/chmod
0.5sprocessexec /usr/local/bin/chmod
0.5sprocessexec /usr/sbin/chmod
0.5sprocessexec /usr/bin/chmod
0.5sprocessexec /usr/bin/python3.11
0.5sprocessexited 2: [anzhuang statesync] 找不到共享模块 cad_statesync_core.py(试过 CAD_FILE_READER_SCRIPTS / 融合包 modules / 技能根)
0.5sprocessbundled script run in the sandbox with no arguments
0.6sprocessexited 2: contract_guard.py: error: 需要契约路径,或用 --self-test
0.6sprocessbundled script run in the sandbox with no arguments
0.7sprocessbundled script run in the sandbox with no arguments
0.7sprocessexited 23: 正式链路:cad-file-reader 产出 cad-contract/v1 → lis-skill-bus → mep_scan --contract。
0.7sprocessbundled script run in the sandbox with no arguments
0.8sfilesystemread ~/.workbuddy/skills/ outside project
0.8snetworkCONNECT tcp://github.com:443
0.8snetworkCONNECT tcp://release-assets.githubusercontent.com:443 · not in the allow-list, refused
0.8sprocessexec /usr/bin/dirname
0.8sprocessexec /usr/bin/mktemp
0.8sprocessexec /usr/bin/curl
0.8sprocessexec /usr/bin/rm
0.8sprocessexited 3: [安装算量] 下载底座失败。离线请设 SKILL_NO_BASE=1 降级使用;联网重试建议 export CAD_FILE_READER_VERSION=0.34.1 后重试。
0.8sprocessbundled script run in the sandbox with no arguments
0.9sprocessbundled script run in the sandbox with no arguments
0.9sprocess8 of 19 bundled scripts run with no arguments (the rest were not run: per-bundle cap)
0.9sprocessscript backend runs setup only
0.9sprocessscript backend runs setup only
1.0sprocesssecond launch, clock +45 days, CI unset
1.2sfilesystemread ~/.workbuddy/skills/ outside project
1.2snetworkCONNECT tcp://github.com:443
1.2snetworkCONNECT tcp://release-assets.githubusercontent.com:443 · not in the allow-list, refused
1.2sprocessexec /usr/bin/dirname
1.2sprocessexec /usr/bin/find
1.2sprocessexec /usr/bin/mktemp
1.2sprocessexec /usr/bin/curl
1.2sprocessexec /usr/bin/rm
1.2sprocessexited 3: [安装算量] 下载底座失败。离线请设 SKILL_NO_BASE=1 降级使用;联网重试建议 export CAD_FILE_READER_VERSION=0.34.1 后重试。

Static findings

MEDIUM
File permission modification commandscripts/runtime_env.sh:18 · File permission modification command — agent changes file ownership/permissionsfind "$SKILL_DIR/scripts" -maxdepth 1 -type f \( -name '*.sh' -o -name '*.py' \) ! -perm -u+x -exec chmod +x {} + 2>/dev/null || trueFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
File permission modification commandscripts/setup.sh:6 · File permission modification command — agent changes file ownership/permissionsfind "$SKILL_DIR/scripts" -maxdepth 1 -type f \( -name '*.sh' -o -name '*.py' \) ! -perm -u+x -exec chmod +x {} + 2>/dev/null || trueFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Python subprocess execution detectedscripts/mep_plan.py:1123 · Python subprocess execution detectedr = subprocess.run(["bash", str(validator), path], capture_output=True, text=True, timeout=180)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedscripts/lint_cross_platform.py:105 · Python subprocess execution detected# 仅对命令类规则,在 .py 里只追 subprocess.run([...]) 这样的真实调用。Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Privileged sudo commandscripts/lint_cross_platform.py:67 · Privileged sudo command — agent may execute with elevated privileges"dylib 是 macOS 原生库格式。fix: 仓库内 sudo 二进制请走 platform 分支;Windows/Linux 由 pip wheel 提供"),Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05

Cost to the agent

Instructions loaded
≈861 tok
1 file · 40 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.

What would make this Verified

24 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Contacts a host you did not declare≈ +23 on the totalList every host you talk to in SKILL.md frontmatter (`network: [api.example.com]`), or stop contacting the ones you cannot justify.
  2. Unclaimed publisher≈ +8 on the totalClaim lisong2003-lgtm: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  3. Touches files outside the project≈ +7 on the totalKeep reads and writes inside the working directory, or declare the exact paths in `filesystem:` and explain why in the README.
  4. Privileged sudo command≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
  5. File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
  6. Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
Publish a new version to ClawHub, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim lisong2003-lgtmPropose a patch to the maintainer