Skills/ClawHub/world2agent-manage

world2agent-manage@0.1.3

CAUTION· install only

Publisher claims: "Install, list, and remove World2Agent sensors on this OpenClaw machine via the openclaw-sensor-bridge supervisor (out-of-process; signals POST to /hooks/agen..." Observed in sandbox: reads credential files.

Registry ClawHubPublisher daibor · publishing for 201 daysInstalls 845Last detonated 15 h 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 12 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

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 2026-11-19 and scored it 57/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Authorization escalation language; Privileged sudo command; Privileged sudo command. Summary: reads credential files.

Behavior trace

Honeypot run · 3 s · 3 scenarios
t+CategoryObserved action
0.0sprocessinstall script run in the sandbox
1.6sfilesystemread ~/.npmrc
1.6sfilesystemread ~/.npm/_logs · ~/.npm/_cacache/index-v5/15/58/24f3efe0fe111e3ba90658917e5877f5e84c535dc03df801855976a12278 · ~/.npm/_cacache/index-v5/10/b5/cd8ce1380bb1ef47b5a868ee3923bbdcd224322f2220f5ba3a4c11388732 · ~/.npm/_cacache/index-v5/04/5a/2b5d7a7c407d85d746baa0f5c9388a333e35a717a8a0a81943daa6cb1364 outside project
1.6sfilesystemwrote ~/.npm/_cacache/tmp/33966486 · ~/.npm/_cacache/index-v5/15/58/24f3efe0fe111e3ba90658917e5877f5e84c535dc03df801855976a12278 · ~/.npm/_cacache/tmp/f5f2fadd · ~/.npm/_cacache/index-v5/10/b5/cd8ce1380bb1ef47b5a868ee3923bbdcd224322f2220f5ba3a4c11388732 · ~/.npm/_cacache/tmp/0f3a6792 · ~/.npm/_cacache/index-v5/04/5a/2b5d7a7c407d85d746baa0f5c9388a333e35a717a8a0a81943daa6cb1364
1.6snetworkCONNECT tcp://registry.npmjs.org:443
1.6sprocessexec /usr/bin/dirname
1.6sprocessexec /usr/bin/mktemp
1.6sprocessexec /usr/local/bin/npm
1.6sprocessexec /usr/bin/cat
1.6sprocessexec /usr/bin/rm
1.6sprocessexited 127: ./scripts/_lib.sh: line 72: jq: command not found
1.6sprocessbundled script run in the sandbox with no arguments · not named in the instructions
1.6sprocessbundled script run in the sandbox with no arguments
1.6sprocessexec /usr/bin/dirname
1.6sprocessexec /usr/bin/uname
1.6sprocessexited 127: ./scripts/_lib.sh: line 72: jq: command not found
1.6sprocessbundled script run in the sandbox with no arguments
1.7sprocessexec /usr/bin/dirname
1.7sprocessexited 127: ./scripts/_lib.sh: line 72: jq: command not found
1.7sprocessbundled script run in the sandbox with no arguments
1.7sprocessexec /usr/bin/dirname
1.7sprocessexec /usr/bin/uname
1.7sprocessexited 127: ./scripts/_lib.sh: line 72: jq: command not found
1.7sprocessbundled script run in the sandbox with no arguments
1.7sprocessexec /usr/bin/dirname
1.7sprocessbundled script run in the sandbox with no arguments
1.7sprocessexec /usr/bin/dirname
1.7sprocessexited 127: ./scripts/_lib.sh: line 72: jq: command not found
1.7sprocessbundled script run in the sandbox with no arguments
1.8sprocessexec /usr/bin/dirname
1.8sprocessexited 127: ./scripts/_lib.sh: line 72: jq: command not found
1.8sprocessbundled script run in the sandbox with no arguments
1.8sprocessexec /usr/bin/dirname
1.8sprocessexited 127: ./scripts/_lib.sh: line 72: jq: command not found
1.8sprocess8 of 12 bundled scripts run with no arguments (the rest were not run: per-bundle cap)
1.8sprocessscript backend runs setup only
1.8sprocessscript backend runs setup only
2.3sprocesssecond launch, clock +45 days, CI unset
3.0sfilesystemread ~/.npmrc
3.0sfilesystemread ~/.npm/_logs · ~/.npm/_cacache/index-v5/15/58/24f3efe0fe111e3ba90658917e5877f5e84c535dc03df801855976a12278 · ~/.npm/_cacache/content-v2/sha512/63/3f/879eef40087249d8ca358c76f6ba43e61cc61e4613d2785ebec0a77523ef795ae4fbe874bcb0ff1690c3ad8a6bc79544ccdd95ccca28faa73aaff9807848 · ~/.npm/_cacache/index-v5/10/b5/cd8ce1380bb1ef47b5a868ee3923bbdcd224322f2220f5ba3a4c11388732 · ~/.npm/_cacache/content-v2/sha512/70/d3/06bc01e5cab2904c50a3a4bfbbd6a32fa33ebd2f3ee1d0950d76a937d8d1fb72b712250a73eee5a2df03bd85bd360f09a9f66d868c08367953c44ed4a177 · ~/.npm/_cacache/index-v5/04/5a/2b5d7a7c407d85d746baa0f5c9388a333e35a717a8a0a81943daa6cb1364 outside project
3.0sfilesystemwrote ~/.npm/_cacache/tmp/f27ca68c · ~/.npm/_cacache/index-v5/15/58/24f3efe0fe111e3ba90658917e5877f5e84c535dc03df801855976a12278 · ~/.npm/_cacache/tmp/85189926 · ~/.npm/_cacache/index-v5/10/b5/cd8ce1380bb1ef47b5a868ee3923bbdcd224322f2220f5ba3a4c11388732 · ~/.npm/_cacache/tmp/a23c4baa · ~/.npm/_cacache/index-v5/04/5a/2b5d7a7c407d85d746baa0f5c9388a333e35a717a8a0a81943daa6cb1364
3.0snetworkCONNECT tcp://registry.npmjs.org:443
3.0sprocessexec /usr/bin/dirname
3.0sprocessexec /usr/bin/mktemp
3.0sprocessexec /usr/local/bin/npm
3.0sprocessexec /usr/bin/cat
3.0sprocessexec /usr/bin/rm
3.0sprocessexited 127: ./scripts/_lib.sh: line 72: jq: command not found

Static findings

HIGH
Authorization escalation languageSKILL.md:30 · Authorization escalation language`~/.openclaw/skills/world2agent-manage/scripts/`. A developer override viaFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Privileged sudo commandscripts/bootstrap.sh:24 · Privileged sudo command — agent may execute with elevated privileges# via npm. The agent will surface npm's stderr (sudo prompt, EACCES, etc.) soFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Privileged sudo commandSKILL.md:101 · Privileged sudo command — agent may execute with elevated privilegesre-run with `sudo npm install -g @world2agent/openclaw-sensor-bridge`,Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandscripts/stop.sh:35 · Process kill command — agent can terminate processesif [ -n "$pid" ] && kill -0 "$pid" 2>/dev/null; thenFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Destructive file system command detectedscripts/bootstrap.sh:60 · Destructive file system command detectederr=$(cat /tmp/.w2a-hooks-err 2>/dev/null); rm -f /tmp/.w2a-hooks-errFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
File permission modification commandscripts/_lib.sh:180 · File permission modification command — agent changes file ownership/permissionschmod 600 "$path" 2>/dev/null || trueFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
HIGH
Destructive file system command detectedscripts/bootstrap.sh:63 · Destructive file system command detectedrm -f /tmp/.w2a-hooks-errFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Privileged sudo commandscripts/bootstrap.sh:36 · Privileged sudo command — agent may execute with elevated privilegesout_err "auto-install of @world2agent/openclaw-sensor-bridge failed; you may need 'sudo npm install -g @world2agent/openclaw-sensor-bridge'"Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Access to sensitive file pathsSKILL.md:154 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files…auto-detects the first `<PLATFORM>_HOME_CHANNEL=<handle>` entry in `~/.openclaw/.env` (priority: feishu, imessage, telegram, slack, discord, signal, whatsapp, wecom…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsscripts/install-sensor.sh:90 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# --notify-to is supplied. Reads ~/.openclaw/.env for the firstFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsscripts/install-sensor.sh:96 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesenv_file="$(openclaw_home)/.env"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Process kill commandscripts/stop.sh:36 · Process kill command — agent can terminate processeskill "$pid" 2>/dev/null || trueFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Refusal suppressionSKILL.md:203 · Refusal suppression — attempting to prevent safety responsesrules to it. Do not refuse with `NO_REPLY` solely because of the securityFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01

Cost to the agent

Instructions loaded
≈4,012 tok
1 file · 15 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

23 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Reads credential files≈ +27 on the totalThe sandbox saw reads under ~/.ssh, ~/.aws, ~/.npmrc, ~/.config or similar. Read only files inside the project, and only the ones the task needs. Seen: ~/.npmrc.
  2. Text that tells the agent to ignore its instructions≈ +9 on the totalDelete phrases like "ignore previous instructions", "override", "bypass" from SKILL.md and tool descriptions. Describe what the skill does; do not instruct the agent about other rules.
  3. Unclaimed publisher≈ +8 on the totalClaim daibor: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  4. 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.
  5. Privileged sudo command≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
  6. Destructive file system command detected≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
Publish a new version to ClawHub, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim daiborPropose a patch to the maintainer