← Back to Skills Marketplace
318
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install yan-watchman
Description
yan-watchman is a Rust-based multi-platform AI assistant message watchdog with intelligent importance scoring and adaptive monitoring frequency.
Usage Guidance
Do not run the SKILL.md commands on a production machine. The package contains only the instruction file but tells the agent to create and build a Rust project that does not exist here and to 'publish' it; that's an incoherence. Before installing, ask the publisher for: (1) a source repository or release artifact (GitHub/official host) containing the claimed Rust source or a prebuilt binary, (2) clear explanations of what credentials (Discord/Feishu tokens) the skill needs and why, and (3) a reproducible install spec (preferred: a signed release or well-known package). If you must test, do so in a sandboxed VM or container, inspect any files the skill would create, and avoid providing sensitive tokens until you can review source code and provenance.
Capability Analysis
Type: OpenClaw Skill
Name: yan-watchman
Version: 1.0.0
The skill uses manipulative prompt injection techniques, employing 'Command Seal' (令咒) and 'Highest Command' rhetoric to compel the agent to execute instructions immediately and bypass normal reasoning. While the provided shell commands in SKILL.md are limited to local file operations and 'cargo build' for a Rust project, the use of persona-driven overrides to ensure 'self-evolution' and the absence of the actual source code (src/*.rs) for the multi-platform monitoring logic make the bundle's intent unverifiable and suspicious.
Capability Assessment
Purpose & Capability
The SKILL.md repeatedly describes a Rust project (Cargo.toml, src/, SQLite storage, tokio) and lists rust/cargo/sqlite3 as requirements, but the registry metadata shows no required binaries, no source repo, and the package contains only SKILL.md. A Rust-based multi-platform monitor would reasonably require source files or a binary plus API credentials for monitored platforms; those are missing.
Instruction Scope
Runtime instructions tell the agent to create a full project tree, run `cargo build --release`, copy src/config/README, and write a new SKILL.md to a publish directory. However, the package contains no source files to copy or build. The instructions also claim multi-platform monitoring (Discord/Feishu/WebChat) but do not direct the agent to obtain or use any platform credentials, nor do they justify reading other system files — this is scope-incoherent.
Install Mechanism
There is no declared install spec (the skill is instruction-only). The SKILL.md expects a local build with cargo; that implies installing/using developer toolchain, but no source is included. While no external download URL is present (lower install risk), the build instructions as provided are inconsistent with the actual package contents.
Credentials
The skill claims to monitor external chat platforms but requires no API keys, tokens, or config paths in the registry metadata. The SKILL.md's internal YAML mentions rust/cargo/sqlite3 but the registry lists none. Requesting no credentials for a multi-platform monitor is disproportionate and unexplained.
Persistence & Privilege
The skill does not request always:true, does not declare system-wide modifications, and its instructions operate within ~/.openclaw/workspace/skills which is the skill area. That said, the instructions will write files and run builds in the user's home workspace — actions that should be run only with explicit consent and in an appropriate environment.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install yan-watchman - After installation, invoke the skill by name or use
/yan-watchman - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release: Multi-platform AI assistant message guardian system with intelligent importance scoring
Metadata
Frequently Asked Questions
What is yan-watchman?
yan-watchman is a Rust-based multi-platform AI assistant message watchdog with intelligent importance scoring and adaptive monitoring frequency. It is an AI Agent Skill for Claude Code / OpenClaw, with 318 downloads so far.
How do I install yan-watchman?
Run "/install yan-watchman" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is yan-watchman free?
Yes, yan-watchman is completely free (open-source). You can download, install and use it at no cost.
Which platforms does yan-watchman support?
yan-watchman is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created yan-watchman?
It is built and maintained by eightroad (@eightroad); the current version is v1.0.0.
More Skills