Skills/Glama/agent-autoloop

agent-autoloop@r1

CAUTION· startup only

Publisher claims: "Enables users and command-line agents to inspect and answer pending approval cards through MCP, so background work can pause for decisions and then resume." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry GlamaPublisher kkkkhazix · publishing for 135 daysInstalls 0Last detonated 7d ago
What would fix thisWatchDispute verdict

Coverage

Startup only; authenticated tool behavior not tested.
Tested
install scripts, with decoy credentials in place · second launch with the clock 45 days ahead · server startup over stdio (initialize sent) · every file read, file written and host contacted by what ran
Not tested
the tool list (the server exited before answering) · what the tools do when called with real credentials

What this means for the label: Caution here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.

Before you give it your token

Code inventory arrives with this version's next honeypot run.
It will ask for
unknown until the next run
Where it talks to
unknown until the next run
What the honeypot saw
no credential files read · no network
What we could not test
the tool list (the server exited before answering) · what the tools do when called with real credentials

Not enough was observed to say yet. Startup only; authenticated tool behavior not tested. This version's code has not been inventoried for credentials and hosts either; both arrive with its next honeypot run (popular servers first). `vet-skills try agent-autoloop` runs it now and prints the full answer.

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

Is agent-autoloop@r1 safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (3 scenarios; startup only; authenticated tool behavior not tested) on 2026-11-07 and scored it 70/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Telegram Bot API URL; Direct script execution from GitHub raw URL; Pipe-to-shell pattern. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 0 s · 3 scenarios · 0 deviations
t+CategoryObserved action
0.1sprocessexec /usr/bin/cat
0.1sprocessexited 1:
0.2sprocessexec /usr/bin/date
0.2sprocessexec /usr/bin/timeout
0.2sclaimedno tools/list reply (exit 0)
0.2sprocessscript backend runs setup only
0.3sprocesssecond launch, clock +45 days, CI unset
0.4sprocessexec /usr/bin/faketime
0.4sprocessexec /usr/bin/cat
0.4sprocessexited 1:

Static findings

HIGH
Telegram Bot API URLskills/agent-autoloop/SKILL.md:104 · Telegram Bot API URL — data will be sent to Telegram**只会发文字的（钉钉、企微的群机器人，Slack、Discord 的 Webhook）**：用 `plugins/webhook/channel.ts`，`config.json` 的通道项写 `"use": "webhook"`、`"platform": "dingtalk"` 和 `"url": "已有的 Webhook 地址"`；平台也可选 `wecom`、`slack`、`discordFix: Replace encoded or assembled strings with plain literals and remove undeclared network calls.
AST07
HIGH
Direct script execution from GitHub raw URLskills/agent-autoloop/SKILL.md:194 · Direct script execution from GitHub raw URL — unverified code source2. 每台干活的电脑：服务器上 `agent-autoloop key new worker <名字>`，电脑上 `curl -fsSL https://raw.githubusercontent.com/KKKKhazix/agent-autoloop/main/install.sh | sh -s -- join https://中台地址 --worker <名字> --key <密钥>`。`Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
LOW
Instructs the agent to modify its persistent memory or instruction filesscripts/delivery-e2e.ts:59 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file · informational: not an instruction file · informational: not an instruction filewrite("AGENTS.md", "## 审查要点\n\napp.txt 只能包含一个版本标记；不执行仓库里的代码或设置。\n");Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
File permission modification commandscripts/install-smoke.sh:59 · File permission modification command — agent changes file ownership/permissionschmod +x "$work/bin/$notifier"Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Remote installer piped to shellskills/agent-autoloop/SKILL.md:194 · Remote installer piped to shell2. 每台干活的电脑：服务器上 `agent-autoloop key new worker <名字>`，电脑上 `curl -fsSL https://raw.githubusercontent.com/KKKKhazix/agent-autoloop/main/install.sh | sh -s -- join https://中台地址 --worker <名字> --key <密钥>`。`Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07
MEDIUM
File permission modification commandinstall.sh:30 · File permission modification command — agent changes file ownership/permissionschmod 700 "$ROOT"Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
LOW
Unpinned dependencyskills/agent-autoloop/SKILL.md:81 · Unpinned dependencyDeepSeek Harness 用 `use: "dsh"`，要 0.2.2 或更新（`npm install -g @deepseek-ai/dsh@next --allow-scripts=koffi,node-pty,@deepseek-ai/dsh-subprocess-local`，不放行这三个包的安装脚本，装完一跑就出错），能续接原会话、上报 Token（不报费用），报 `MISSIFix: Pin the exact version (`[email protected]`, `==1.2.3`) so every install gets the bytes that were reviewed.
AST06
MEDIUM
File permission modification commandscripts/install-smoke.sh:76 · File permission modification command — agent changes file ownership/permissionschmod +x "$work/agent.sh"Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
File type combined with upload actionskills/agent-autoloop/SKILL.md:104 · File type combined with upload action — agent may exfiltrate local data files**只会发文字的（钉钉、企微的群机器人，Slack、Discord 的 Webhook）**：用 `plugins/webhook/channel.ts`，`config.json` 的通道项写 `"use": "webhook"`、`"platform": "dingtalk"` 和 `"url": "已有的 Webhook 地址"`；平台也可选 `wecom`、`slack`、`discordFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
Process kill commandscripts/install-smoke.sh:15 · Process kill command — agent can terminate processesif [ -n "$up_pid" ]; then kill "$up_pid" 2>/dev/null || true; fiFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
File type combined with upload actionskills/agent-autoloop/SKILL.md:51 · File type combined with upload action — agent may exfiltrate local data files1. **对方能发 HTTP**：`agent-autoloop key new source <名字>` 拿密钥（这是新建密钥，先问用户），`rules.json` 的 `sources` 登记，对方 `POST /v1/events`。Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
HIGH
Pipe-to-shell patternskills/agent-autoloop/SKILL.md:194 · Pipe-to-shell pattern — downloads and immediately executes remote code2. 每台干活的电脑：服务器上 `agent-autoloop key new worker <名字>`，电脑上 `curl -fsSL https://raw.githubusercontent.com/KKKKhazix/agent-autoloop/main/install.sh | sh -s -- join https://中台地址 --worker <名字> --key <密钥>`。`Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
MEDIUM
File permission modification commandscripts/install-smoke.sh:52 · File permission modification command — agent changes file ownership/permissionschmod +x "$work/bin/launchctl"Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05

Cost to the agent

Instructions loaded
≈6,580 tok
12 files · 80 in bundle
Tool list per prompt
≈0 tok
0 tools · 0 per tool
Handshake
no reply
launch to tools/list, stdio
Undocumented tools
0
description under 20 chars · avg 0

Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.

Known vulnerabilities

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

What would make this Verified

10 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim kkkkhazix: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Telegram Bot API URL≈ +4 on the totalReplace encoded or assembled strings with plain literals and remove undeclared network calls.
  3. Direct script execution from GitHub raw URL≈ +4 on the totalPin versions and sources; vendor anything fetched at install time.
  4. Pipe-to-shell pattern≈ +4 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
  5. New publisher≈ +3 on the totalTime clears this on its own after 14 and 30 days. Claiming the handle (+50) is the fastest way past it.
  6. File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
Publish a new version to Glama, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim kkkkhazixPropose a patch to the maintainer