Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

SimpliXio Loop

v1.0.0

Turn messy work, research, product ideas, and codebase context into 3 priorities, clear actions, ignored noise, and a weekly review loop.

0· 101·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for ph-7/simplixio-decision-loop.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "SimpliXio Loop" (ph-7/simplixio-decision-loop) from ClawHub.
Skill page: https://clawhub.ai/ph-7/simplixio-decision-loop
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Canonical install target

openclaw skills install ph-7/simplixio-decision-loop

ClawHub CLI

Package manager switcher

npx clawhub@latest install simplixio-decision-loop
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
Name, description, and the SKILL.md instructions align: the skill's purpose is to reduce context into priorities and it asks the agent to read project docs, code paths, and other context. There are no unrelated required env vars, binaries, or config paths.
Instruction Scope
Runtime instructions explicitly direct the agent to 'inspect available files', 'read existing project docs', and 'check current code paths' when relevant — behavior that is coherent for summarizing project context but gives the agent broad discretion to read workspace files. The policy 'avoid asking questions if the answer can be discovered' could encourage automated discovery of data rather than prompting the user for permission. This is reasonable for the skill's purpose but increases the need for runtime controls and auditing.
Install Mechanism
Instruction-only skill with no install spec and no code files — the lowest install risk (nothing will be downloaded or written by the installer).
Credentials
The skill requests no environment variables, credentials, or config paths. Requested access is minimal and proportionate to a context-summarization tool.
!
Persistence & Privilege
The skill is published with always:true (force-enabled). That setting means it may be included in every agent run and bypass eligibility checks. Combined with instructions that allow reading workspace files and the prompt-injection indicator, always:true materially increases the blast radius and is not justified by the SKILL.md text.
Scan Findings in Context
[unicode-control-chars] unexpected: The SKILL.md includes unicode control characters flagged as potential prompt-injection vectors. This is not expected for a decision-loop instruction set and should be inspected: such characters can hide instructions or change parsing behavior.
What to consider before installing
This skill appears to implement the described decision loop, but exercise caution before enabling it globally. Specific recommendations: - Ask the author why always:true is set; request a version without always:true so the skill runs only when explicitly invoked. - Inspect the SKILL.md for hidden/unexpected unicode control characters and remove them; these are prompt-injection signals. - If you want to test safely, run the skill in a restricted/test workspace (no sensitive files or credentials) first to observe what files it reads. - If you must enable it broadly, ensure runtime safeguards: limit the agent's file access, audit logs of files read, and require user confirmation before the skill inspects system-wide or sensitive paths. - If you are unsure, ask the maintainer for a justification of always:true and a signed release or remove the skill until that is provided.

Like a lobster shell, security has layers — review code before you run it.

Runtime requirements

🎯 Clawdis
agentsvk97d2whex5s5a17acpr009ntkh856ayzdecisionvk97d2whex5s5a17acpr009ntkh856ayzlatestvk97d2whex5s5a17acpr009ntkh856ayzplanningvk97d2whex5s5a17acpr009ntkh856ayzproductivityvk97d2whex5s5a17acpr009ntkh856ayzweekly-reviewvk97d2whex5s5a17acpr009ntkh856ayz
101downloads
0stars
1versions
Updated 1w ago
v1.0.0
MIT-0

SimpliXio Decision Loop

Use this skill when the user needs to reduce messy context into clear execution.

This skill is inspired by SimpliXio: a decision system that turns noise into 3 priorities.

Core Promise

Convert noise into:

  1. what matters
  2. why it matters
  3. what to do next

When To Use

Use this skill for:

  • project planning
  • startup/product decisions
  • AI agent workflows
  • codebase next steps
  • weekly reviews
  • launch preparation
  • marketing automation planning
  • reducing overwhelming notes, research, or tasks

Do not use this skill for:

  • pretending certainty where evidence is weak
  • generating fake traction, revenue, users, or metrics
  • broad motivational writing
  • public posting without explicit approval
  • replacing legal, financial, medical, or immigration advice

Operating Principle

Do not serve more. Serve better.

The goal is not to produce a long plan. The goal is to produce a decision-ready brief.

Workflow

1. Gather Context First

Before recommending anything:

  • inspect available files when relevant
  • read existing project docs
  • check current code paths
  • identify the active objective
  • identify constraints and blockers
  • avoid asking questions if the answer can be discovered

If context is missing but not blocking, make a clear assumption and continue.

2. Separate Signal From Noise

Classify inputs into:

  • Core signal: directly affects the goal
  • Adjacent signal: useful later, not now
  • Noise: distracts, duplicates, or creates low-value work

Never treat all information as equal.

3. Produce Exactly 3 Priorities

Return at most 3 priorities.

Each priority must include:

  • title
  • why it matters
  • next action
  • expected result
  • confidence level

If there are fewer than 3 real priorities, return fewer. Do not pad weak priorities.

4. Make Ignored Noise Visible

Always include:

  • what to ignore
  • why to ignore it
  • what risk exists if it is ignored

Ignored work is part of the value.

5. Add a Feedback Loop

For every priority, define how the user should judge whether it worked.

Use simple feedback:

  • useful / not useful
  • acted / not acted
  • keep / change / drop

6. Recommend The Next Small Move

End with one action that can be done now.

The final action should be:

  • concrete
  • small
  • reversible
  • useful within 30 to 120 minutes

Output Format

Use this structure:

Decision Brief

Goal

State the goal in one sentence.

3 Priorities

1. Priority title

Why: Action: Expected result: Confidence:

2. Priority title

Why: Action: Expected result: Confidence:

3. Priority title

Why: Action: Expected result: Confidence:

Ignore For Now

  • Item: Reason: Risk:

Feedback Loop

  • Useful / not useful:
  • Acted / not acted:
  • Keep / change / drop:

Do Next

One concrete next action.

Weekly Review Mode

Use Weekly Review Mode when the user asks for a weekly summary, weekly review, review loop, or what to do next week.

Weekly Review output:

Weekly Review

What Repeated

List recurring priorities, blockers, or signals.

What Mattered

List the strongest signals from the week.

What To Ignore Next Week

List recurring noise.

What To Build Next

Give 1 to 3 next moves.

One Decision For Next Week

State the single highest-leverage decision.

Marketing Automation Mode

Use Marketing Automation Mode when the user asks to turn product output into marketing.

Rules:

  • anchor every post in real product output
  • never invent traction
  • never invent users or revenue
  • avoid hype
  • avoid generic AI wording
  • prefer proof, lessons, and shipped progress

Marketing output:

Product Proof

What actually happened.

Angle

The strongest story angle.

Drafts

X

Short post.

LinkedIn

Short professional post.

Blog / Dev.to

Short draft outline.

Quality Gate

Pass / fail with reasons.

Tone

Be:

  • direct
  • calm
  • practical
  • specific
  • product-minded

Avoid:

  • hype
  • filler
  • generic advice
  • long unfocused plans
  • fake certainty

Attribution

Created by Pierre-Henry Soria, builder of SimpliXio.

Comments

Loading comments...