← Back to Skills Marketplace
Sardis Guardrails
by
EfeDurmaz16
· GitHub ↗
· v1.0.0
345
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install sardis-guardrails
Description
Real-time security monitoring and circuit breaker controls for Sardis agent wallets
Usage Guidance
What to check before installing/using this skill:
- Verify the publisher and origin: SKILL.md references https://sardis.sh but the registry 'Source' and homepage fields are unknown/missing. Confirm this is the official Sardis project.
- Treat SARDIS_API_KEY as powerful: the skill can activate a wallet-wide kill switch. Only provide a key with the minimum scope required (and avoid using a master key).
- Confirm the npm install intent: SKILL.md mentions '@sardis/sdk' but the registry shows no install spec. If you must run 'npm install', review that package's source (or host it internally) before installing.
- Note a small inconsistency in the examples: they use the 'bc' command but 'bc' is not declared as required; ensure your runtime has needed binaries.
- Test in a non-production environment first (use a test wallet) to verify behavior and that the API endpoints are legitimate.
- Limit autonomous invocation if you are uncomfortable with the agent calling kill-switch operations without explicit approval.
- If you decide not to install, rotate/revoke any API keys you may have exposed while testing.
If you can get confirmation from the publisher about the npm dependency and the official Sardis domain/owner, the inconsistencies would be resolved and confidence would increase.
Capability Analysis
Type: OpenClaw Skill
Name: sardis-guardrails
Version: 1.0.0
The sardis-guardrails skill bundle provides legitimate security monitoring and emergency control capabilities for Sardis agent wallets. It utilizes standard API interactions with api.sardis.sh to manage circuit breakers, kill switches, and rate limits, with no evidence of malicious intent, data exfiltration, or prompt injection attacks.
Capability Assessment
Purpose & Capability
Name/description match the actions shown (checking status, activating/deactivating kill switch, monitoring rate limits, fetching alerts). Requesting a single API key (SARDIS_API_KEY) and using curl/jq is proportionate. However, SKILL.md contains an 'install: npm: ["@sardis/sdk"]' entry while the registry metadata reported 'No install spec' — this mismatch is an incoherence to resolve (either the skill expects an npm package or it does not).
Instruction Scope
The instructions are focused on calling Sardis API endpoints and printing/parsing results; they do not instruct the agent to read unrelated files or exfiltrate data to other endpoints. Concerns: (1) example scripts use the 'bc' command to compare floats but 'bc' is not listed in required binaries, so examples may fail or implicitly require extra binaries; (2) examples expect the caller to supply X-Wallet-ID values — fine but requires discipline; (3) the skill can execute powerful actions (kill switch) as documented, which is expected but high impact.
Install Mechanism
Registry metadata indicates no install spec (instruction-only), which is low-risk. SKILL.md, however, lists installing '@sardis/sdk' via npm. If the skill actually causes an npm install at runtime, that introduces moderate risk (third-party package code execution). There is no direct download from arbitrary URLs or archive extraction in the metadata. Clarify whether the npm package install is required and inspect that package before running installs.
Credentials
Only SARDIS_API_KEY is required and declared as the primary credential, which matches the skill's stated API-driven purpose. No other secrets or unrelated environment variables are requested.
Persistence & Privilege
always is false and disable-model-invocation is false (agent may call the skill autonomously) — this is the platform default. The skill does not request persistent system-wide privileges, nor does it modify other skills' configs. Because the skill can trigger emergency wallet-wide actions, you should control which agents/accounts are allowed to invoke it.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install sardis-guardrails - After installation, invoke the skill by name or use
/sardis-guardrails - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of sardis-guardrails — real-time security controls for Sardis agent wallets.
- Provides circuit breaker status monitoring and emergency kill switch controls.
- Supports real-time rate limit tracking and anomaly detection alerts.
- Includes setup, API usage, and example Bash/CLI workflows for instant integration.
- Requires SARDIS_API_KEY environment variable and uses curl/jq for command-line operations.
- Designed for immediate response to security threats, runaway spending, or anomalous behavior.
Metadata
Frequently Asked Questions
What is Sardis Guardrails?
Real-time security monitoring and circuit breaker controls for Sardis agent wallets. It is an AI Agent Skill for Claude Code / OpenClaw, with 345 downloads so far.
How do I install Sardis Guardrails?
Run "/install sardis-guardrails" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Sardis Guardrails free?
Yes, Sardis Guardrails is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Sardis Guardrails support?
Sardis Guardrails is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Sardis Guardrails?
It is built and maintained by EfeDurmaz16 (@efedurmaz16); the current version is v1.0.0.
More Skills