Agent Andri

Agent Andri periodically sends its status report to the meeting-room by appending a message to a designated file.

Security Scan

The skill mostly does what it says (periodically appends status messages to a file), but there are mismatches between the documentation and the script and the script's infinite behavior could cause resource or operational issues.

Purpose & Capability

Name/description match the actual behavior: the included script appends status lines to ~/.openclaw/workspace/skills/meeting-room/to_leader.txt every 30 seconds. No unrelated credentials, binaries, or external endpoints are requested.

Instruction Scope

SKILL.md says the script reads AGENT_NAME, NV_API_KEY and MODEL; the shipped script does not read those environment variables and instead hardcodes AGENT_NAME="Andri". The script runs an infinite loop that continuously appends to the file every 30s, which is broader operational behavior than a one-off status send.

Install Mechanism

No install spec; code is instruction-only with a single shell script. Nothing is downloaded or installed by the skill itself.

Credentials

The SKILL.md documents environment variables (AGENT_NAME, NV_API_KEY, MODEL) but the script does not consume them. The skill declares no required env vars, which is coherent, but the mismatch may be accidental or sloppy documentation.

Persistence & Privilege

The script is an infinite loop that appends to a file every 30 seconds; if the agent runs this unmonitored it will continue indefinitely, potentially filling disk or spamming the meeting-room file. The skill is not configured 'always:true', but autonomous invocation could still start the script.

What to consider before installing

This skill's behavior is simple and mostly aligned with its description, but note three issues before installing: (1) the script hardcodes AGENT_NAME instead of using advertised env vars—if you expect dynamic names update the script to read $AGENT_NAME; (2) it runs an infinite loop appending every 30s—ensure you have a plan to start/stop it, limit file growth (log rotation or max size), and monitor CPU/disk use; (3) the shebang is Termux-specific (/data/.../bash) so it may fail on non-Android systems—adjust the shebang to a standard shell (/usr/bin/env bash). If you proceed, run the script under a supervisor or with a wrapper that enforces lifetime and log rotation, and inspect the target file and permissions to ensure no sensitive data is being exposed.

Skill ini merepresentasikan agen pekerja Andri. Agen secara periodik mengirimkan statusnya ke meeting‑room dengan menuliskan ke file ~/ .openclaw/workspace/skills/meeting-room/to_leader.txt.

Variabel lingkungan

• AGENT_NAME – Nama agen (di‑set otomatis oleh wrapper).
• NV_API_KEY – API‑key khusus agen (tidak dipakai di contoh ini, hanya disimpan bila diperlukan).
• MODEL – Model AI yang dipakai agen.

Skrip utama

scripts/status_report.sh membaca variabel di atas, men‑generate pesan status, dan men‑append‑kan ke file meeting‑room.