← Back to Skills Marketplace
mindbomber

AANA Guardrail Skill Bundled Helper Variant

by mindbomber · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ✓ Security Clean
84
Downloads
1
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install aana-guardrail-bundled
Description
Provides a local Python helper to send minimal review payloads to a trusted AANA HTTP bridge for policy guidance inside the skill package.
README (SKILL.md)

AANA Guardrail Skill Bundled Helper Variant

Use this variant only when the user or administrator wants a bundled, inspectable helper inside the skill package.

This package includes a small Python helper at bin/aana_guardrail_check.py. The helper does not implement AANA policy itself. It only sends a minimal review payload to a separately running AANA HTTP bridge on localhost.

Safety Boundary

  • Do not run the helper unless the user explicitly approves using the configured AANA bridge for this task.
  • Do not send payloads to remote hosts. The bundled helper rejects non-localhost URLs.
  • Do not include API keys, bearer tokens, passwords, full payment numbers, unnecessary account records, or unrelated private messages.
  • Prefer redacted summaries over raw sensitive content.
  • Treat AANA recommendations as advisory unless the bridge and policy are trusted and auditable.
  • If the helper or bridge is unavailable, use manual review.

Review Payload

Use a minimal JSON payload:

{
  "adapter_id": "support_reply",
  "request_summary": "draft a refund support reply",
  "candidate_summary": "reply would promise refund eligibility",
  "evidence_summary": ["refund eligibility is unknown"],
  "allowed_actions": ["accept", "revise", "ask", "defer", "refuse"]
}

Helper Behavior

The helper:

  • reads one JSON payload file,
  • validates that it is a JSON object,
  • blocks obvious secret-like keys,
  • sends the payload to a localhost AANA bridge,
  • prints the JSON result,
  • exits nonzero if the result recommends anything other than accept.

The helper does not:

  • install dependencies,
  • execute other scripts,
  • infer checker paths,
  • contact remote hosts,
  • create event files,
  • delete user files.

Decision Rule

  • If recommended_action is accept, proceed.
  • If recommended_action is revise, revise before acting.
  • If recommended_action is ask, ask the user for missing information.
  • If recommended_action is defer, route to a stronger tool or human review.
  • If recommended_action is refuse, pause, explain the reason, and ask for review when the decision affects important work.

If no trusted bridge is configured, do not use this helper.

Usage Guidance
This skill appears safe and consistent with its stated purpose. Before installing or using it, make sure the localhost AANA bridge is one you trust, keep review payloads short and redacted, and use only the approved bridge URL.
Capability Analysis
Type: OpenClaw Skill Name: aana-guardrail-bundled Version: 1.0.1 The skill bundle provides a helper script for interfacing with a local guardrail service (AANA bridge). It demonstrates strong security practices, including restricting network communication to 'localhost' only, scanning JSON payloads for sensitive keys (e.g., 'api_key', 'password') to prevent accidental leakage, and using only the Python standard library to avoid supply chain risks. The instructions in SKILL.md and the logic in bin/aana_guardrail_check.py are consistent, safety-oriented, and lack any indicators of malicious intent or significant vulnerabilities.
Capability Tags
cryptocan-make-purchasesrequires-sensitive-credentials
Capability Assessment
Purpose & Capability
The stated purpose matches the bundled code: it reads a review JSON file, sends a redacted event to a localhost AANA bridge, and prints the bridge response. The external bridge itself is not included, so its trustworthiness remains outside this package.
Instruction Scope
The instructions explicitly require user approval before using the helper, forbid secrets and unrelated private data in payloads, and state that AANA recommendations are advisory unless the bridge is trusted.
Install Mechanism
There is no install script, no dependency installation, and requirements.txt is intentionally empty. The single Python helper is bundled and readable.
Credentials
The helper makes an HTTP POST to localhost and reads one user-specified payload file, which is proportionate for the guardrail purpose. Capability signals mentioning purchases or credentials are not supported by the file contents, which instead say not to include payment details or secrets.
Persistence & Privilege
The artifacts show no persistence, credential use, background service creation, event-file writing, deletion, or privilege escalation.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install aana-guardrail-bundled
  3. After installation, invoke the skill by name or use /aana-guardrail-bundled
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- Added compiled Python helper file: bin/__pycache__/aana_guardrail_check.cpython-313.pyc. - Updated documentation to clarify the helper sends review payloads only to localhost (removed reference to 127.0.0.1).
v1.0.0
Initial release of the bundled AANA Guardrail helper skill. - Provides a minimal Python helper script to send review payloads to a locally running AANA HTTP bridge. - The helper validates payloads, blocks obvious secret keys, and only contacts localhost addresses. - Printed results include gate recommendations; exits nonzero when the action is not accepted. - Includes example payload, schema, and detailed usage/safety documentation. - No dependencies are installed; remote hosts and sensitive data are strictly excluded by default.
Metadata
Slug aana-guardrail-bundled
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is AANA Guardrail Skill Bundled Helper Variant?

Provides a local Python helper to send minimal review payloads to a trusted AANA HTTP bridge for policy guidance inside the skill package. It is an AI Agent Skill for Claude Code / OpenClaw, with 84 downloads so far.

How do I install AANA Guardrail Skill Bundled Helper Variant?

Run "/install aana-guardrail-bundled" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is AANA Guardrail Skill Bundled Helper Variant free?

Yes, AANA Guardrail Skill Bundled Helper Variant is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does AANA Guardrail Skill Bundled Helper Variant support?

AANA Guardrail Skill Bundled Helper Variant is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created AANA Guardrail Skill Bundled Helper Variant?

It is built and maintained by mindbomber (@mindbomber); the current version is v1.0.1.

💬 Comments