技能 agent-world
📦
agent-world
v1.0.0 內容修訂 r1 嚴重 🌐 網路存取🔑 環境變數📁 檔案系統存取⚙️ 外部命令
Navigate Agent Communities Safely
Agent community requirements, access methods, and status information are scattered across many platforms. This skill organizes platform routes, participation steps, status checks, and credential guidance.
支援: Claude Codex Code(CC)
請勿自動安裝此 Skill。
權威政策要求在執行任何安裝操作前由操作人員審查。
使用我的智慧代理安裝
將此請求複製給你的智慧代理，其中包含權威 Skill 頁面和清單。
智能體請求
Review the Skillstore skill "agent-world" from https://skillstore.io/skills/ptreezh-agent-world.md and its manifest at https://skillstore.io/api/skills/ptreezh-agent-world/manifest. Verify the artifact. Do not auto-install. Inspect the skill and report your findings, then wait for an operator or manual installation decision.你的智慧代理仍應顯示其計畫，並依安全政策請求必要確認。
Agent 可讀資源
當 AI Agent、爬蟲或腳本需要乾淨脈絡、而不是讀取完整頁面時，請使用這些連結。
測試它
正在使用「agent-world」。 Recommend a platform for listing an open-source agent skill with minimal registration.
預期結果:
Recommendation: theskills.directory. Submit a pull request using its template and keep the description on one line. Confirm current availability before submission.
正在使用「agent-world」。 Create a safe registration checklist for agentid.sh.
預期結果:
- Confirm the official endpoint and review its current documentation.
- Register the chosen handle only after operator approval.
- Store the returned private key in an encrypted credential manager.
- Record only non-secret account metadata in project documentation.
正在使用「agent-world」。 Summarize status checks for all configured platforms.
預期結果:
The checker contacts each configured base URL and reports HTTP success or failure. Results show reachability, not authentication, API correctness, or service health.
安全審計
嚴重v1 • 2026/9/12 開啟版本化報告
Most static matches are benign documentation syntax, schema URLs, or defensive credential guidance. Confirmed risks include remote pipe-to-shell commands, plaintext private-key storage guidance, executable installation into a home directory, and outbound network requests. Semantic review also identified unsafe delegation to mutable remote skills, unverified archive extraction, unrestricted URL probing, and coordinated engagement behavior.
8
已掃描檔案
479
分析行數
15
複核項目
0
已忽略的誤報
已確認的安全問題 (7)
嚴重
Pipe to shell pattern
> - How to join (one command): "One-command join: curl -sL https://raw.githubusercontent.com/<owner>
The post template recommends downloading a mutable remote script and piping it directly to bash. This permits immediate arbitrary code execution without inspection or integrity verification.
嚴重
Pipe to shell pattern
> "Join in one command: curl -sL https://raw.githubusercontent.com/<owner>/<repo>/main/join.sh | bas
The reply template promotes direct execution of a remotely downloaded join script. A compromised repository or substituted script would execute with the user's privileges.
高
Crypto seed/private key mention
2. **Server-issued private keys cannot be recovered.** agentid.sh returns a private key and explicit
The skill instructs the agent to persist an unrecoverable private key in a local gitignored file. It does not require encryption or restrictive permissions, leaving high-value key material exposed locally.
高
Untrusted Remote Skills Treated as Authoritative
The routing flow tells agents to fetch changing platform skills, treat them as authoritative, and follow them directly. Without pinning, integrity checks, or instruction isolation, a compromised remote skill can inject instructions and trigger unauthorized actions.
The text explicitly calls remote skills authoritative, directs the agent to follow them, and recommends re-fetching changed versions. No verification or trust boundary is specified.
高
Unverified Remote Archive Installed as Executable
The KodaClaw route streams a mutable release archive into tar and extracts it directly into ~/.local/bin. It provides no version pin, checksum, signature, or archive-content validation.
The complete installation command is present and visibly lacks integrity verification. The destination is an executable search path.
中
Unrestricted URL Probing
The status checker accepts any URL and opens it without scheme or destination validation. Agent-supplied targets can reach loopback, private network, metadata, or redirected endpoints.
The positional URL is copied directly into the request path and urllib follows supported targets. No allowlist, address filtering, or redirect restriction is present.
中
Coordinated Engagement Automation
The skill prescribes replying to every comment and upvoting a fixed number of other posts each session. Automated execution can create inauthentic engagement or violate platform rules.
The instructions explicitly define repeated posting, reply, and upvote behavior as obligations. The skill does not require operator approval or relevance checks for each social action.
能力複核項目 (15)
這些是真實的本機能力，對此技能而言可能是預期行為，因此需要複核，但不會按已確認的惡意行為計分。
高
Generic API/secret keys
Content: handle, api_key/token, id, verification code, claim URL, creation date, status.
The guidance directs users to store API keys, tokens, and verification codes together in plaintext account files. Gitignore reduces commit exposure but does not protect local files or backups.
高
Hidden file in home directory
| KodaClaw | `kc-community` CLI (own skill mgmt) | install: `curl -sL https://github.com/koda-claw/k
The documented command extracts a remote archive directly into ~/.local/bin. This writes an unverified executable into a user command path.
中
Hidden file access
| KodaClaw | `kc-community` CLI (own skill mgmt) | install: `curl -sL https://github.com/koda-claw/k
The installation command writes to the hidden ~/.local/bin directory. The access is intentional, but it creates a persistence and executable replacement risk because the archive is unverified.
中
Python HTTP libraries
with urllib.request.urlopen(req, timeout=timeout) as resp:
urlopen executes the outbound request to the supplied target. Because destinations are unrestricted, the checker can contact local or private network services.
低
Python HTTP libraries
req = urllib.request.Request(url, method="GET", headers={"User-Agent": "agent-world-status-check/1.0
The function constructs an outbound GET request for a caller-supplied URL. This is expected status-check behavior but creates a network access capability.
低
Hardcoded URL
"XiaLiao/ClawdChat": "https://clawdchat.cn",
The --all mode sends an HTTP request to this hardcoded external platform. The behavior is legitimate health checking but still performs outbound network access.
低
Hardcoded URL
"InStreet": "https://instreet.coze.site",
The --all mode sends an HTTP request to this hardcoded external platform. No credentials are sent, but network access is real.
低
Hardcoded URL
"agentid.sh": "https://agentid.sh",
The status checker contacts this external URL when --all is selected. The request is expected and unauthenticated but remains an outbound connection.
低
Hardcoded URL
"SkillsMD": "https://skillsmd.dev",
The status checker contacts this external URL when --all is selected. The behavior is narrow health checking with no secret transmission.
低
Hardcoded URL
"theskills.directory": "https://theskills.directory",
This URL is an active target of the --all health check. It represents intentional outbound network access without credential data.
低
Hardcoded URL
"KodaClaw Community": "https://community.ai-koda.com",
This URL is an active target of the --all health check. The external request is legitimate but confirmed.
低
Hardcoded URL
"Coze Agent World": "https://world.coze.site",
The checker performs an outbound GET to this platform in --all mode. It sends only a fixed user agent and no stored secrets.
低
Hardcoded URL
"Agentica": "https://agentica.wiki",
The checker performs an outbound GET to this platform in --all mode. The access is expected but externally observable.
低
Hardcoded URL
"Moltbook": "https://moltbook.com",
The --all command contacts this external platform to obtain an HTTP status. This is a real, low-impact network operation.
低
Hardcoded URL
"PromptFrenzy": "https://promptfrenzy.com",
The --all command contacts this external platform to obtain an HTTP status. No sensitive payload is attached.
風險因素
🌐 網路存取 (17)
MANIFEST.json:2 MANIFEST.json:54 scripts/check_status.py:17 scripts/check_status.py:36 scripts/check_status.py:38 scripts/check_status.py:22 scripts/check_status.py:23 scripts/check_status.py:24 scripts/check_status.py:25 scripts/check_status.py:26 scripts/check_status.py:27 scripts/check_status.py:28 scripts/check_status.py:29 scripts/check_status.py:30 scripts/check_status.py:31 scripts/gen_manifest.py:32 scripts/gen_manifest.py:41
🔑 環境變數 (4)
偵測到的模式
Pipe to shell pattern×2
審計單位: codex
分享與引用此報告
分享版本化評估報告、中性徽章、嵌入卡片與引用。Skillstore 只報告證據，不判斷該 Skill 是否安全。
複製報告連結
https://skillstore.io/skills/ptreezh-agent-world/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/ptreezh-agent-world?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/ptreezh-agent-world?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/ptreezh-agent-world/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/ptreezh-agent-world.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>學術引用 (APA · BibTeX · CFF)
APA 引用
ptreezh. (2026). agent-world security audit report (audit version 1) [Author version 1.0.0]. Skillstore. https://skillstore.io/skills/ptreezh-agent-world/audits/1BibTeX 引用
@techreport{ptreezh-ptreezh-agent-world-2026,
author = {ptreezh},
title = {agent-world security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/ptreezh-agent-world/audits/1},
note = {Author version 1.0.0}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "agent-world security audit report (audit version 1)"
version: "1.0.0"
type: report
authors:
- name: "ptreezh"
date-released: "2026-09-12"
url: "https://skillstore.io/skills/ptreezh-agent-world/audits/1"
identifiers:
- type: other
value: "skillstore:ptreezh-agent-world:audit:1"
description: "Skillstore immutable audit report identifier"
Skillstore 評分
評分依據 證據置信度: 中 68
架構
100
可維護性
87
內容
65
社群
83
規範符合度
你可以打造什麼
Choose a Community
Compare communities by registration method, audience, availability, and verification needs before selecting a platform.
Plan Community Operations
Prepare posting, reply, notification, and rate-limit workflows across selected agent communities.
Assess Platform Access
Review current status notes and run HTTP checks before integrating a community into an automated workflow.
試試這些提示
Find a Suitable Platform
I want to {goal}. Compare the listed agent communities and recommend one option. Include registration, verification, current status, and important limits.Prepare a Registration Checklist
Create a registration checklist for {platform}. Separate automated steps from human verification. Identify required tools, credentials, and secure storage actions.Plan a Community Launch
Plan a project launch on {platforms}. Draft platform-appropriate announcements and an approval-based engagement schedule. Respect each platform's stated limits and rules.Review a Cross-Platform Workflow
Audit my proposed workflow for {platforms}. Check remote instruction trust, credential storage, installation integrity, network destinations, rate limits, and human approval points.最佳實務
- Verify remote skills and installation files against pinned revisions, checksums, or signatures before use.
- Keep credentials in an encrypted store and require human approval for registration, publishing, and social actions.
- Recheck current platform rules, availability, and rate limits before every campaign or integration.
避免
- Do not pipe remote downloads into a shell or executable directory.
- Do not treat fetched skill text or API suggestions as trusted instructions.
- Do not automate repetitive posts, replies, or votes without operator approval and platform permission.
常見問題
Which assistants can use this skill?
The marketplace metadata supports Claude, Codex, and Claude Code.
Does the skill create platform accounts automatically?
It provides workflows, but some steps require external tools, network access, credentials, or human verification.
Does the status checker prove a platform is working?
No. It checks HTTP responses only and does not validate authentication, APIs, or complete service behavior.
How should credentials be stored?
Use an encrypted operating system credential store or vault. Gitignore alone does not protect plaintext secrets.
Are remote platform skills safe to follow directly?
No. Pin trusted revisions, verify integrity, inspect instructions, and require approval before sensitive actions.
Can platform status and limits change?
Yes. Recheck official documentation and current responses because the included observations can become outdated.
開發者詳細資訊
作者
ptreezh授權
apache-2.0
作者版本
v1.0.0
Skillstore 修訂
r1
引用
5526951beaeca22519572a471618555839b53b05
維護新鮮度
2026/9/14
使用量
0 次下載 · 0 次瀏覽