ClawHub

Mentor Counsel

v1.0.1

Provide strategic guidance with clarity, trade-offs, and risk framing; returns agency to the user without dependency.

⭐ 0· 131·0 current·0 all-time
byMauricio Z.@mzfshark

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for mzfshark/mentor-counsel.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Mentor Counsel" (mzfshark/mentor-counsel) from ClawHub.
Skill page: https://clawhub.ai/mzfshark/mentor-counsel
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

Bare skill slug

openclaw skills install mentor-counsel

ClawHub CLI

Package manager switcher

npx clawhub@latest install mentor-counsel
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (strategic mentoring) match the SKILL.md steps and outputs. The skill asks for no binaries, env vars, or external services — which is appropriate for a purely advisory/decision-framing helper. Minor metadata inconsistencies exist (registry lists version 1.0.1 and owner id kn7411... while bundled files declare version 2.0.0 and owner 'morpheus-agent-001'), but these are bookkeeping issues and do not change the runtime behavior.
✓
Instruction Scope
SKILL.md contains clear, bounded instructions (reframe problem, propose paths, evaluate trade-offs, recommend next actions) and explicit safety rules. It does not instruct the agent to read local files, environment variables, or send data to external endpoints, nor does it grant broad discretionary collection of system context.
✓
Install Mechanism
No install specification and no code files — instruction-only skill. Nothing will be downloaded or written to disk by the skill itself during install, minimizing attack surface.
✓
Credentials
The skill declares no required environment variables, credentials, or config paths and the runtime instructions do not reference any secrets. Requested access is proportionate to an advisory/mentoring role.
✓
Persistence & Privilege
always is false and there is no indication the skill writes system-wide configuration or attempts to persist credentials. Being user-invocable and allowing model invocation is the normal default and appropriate for this skill.
Assessment
This skill appears to be a straightforward, instruction-only mentor/decision-helper and does not request credentials or install code. Before installing, consider: (1) provenance — the registry metadata and embedded files show mismatched owner/version values (harmless but worth noting) and the source/homepage are unknown; if you need strong provenance, ask the publisher for contact or audit history; (2) scope — treat recommendations as advisory (the SKILL.md itself says so); don’t paste sensitive secrets into prompts; (3) verification — for high-impact decisions, validate recommended actions with domain experts or tests. Overall the skill is coherent and low-risk, but maintain normal caution about trusting advice from an unknown source.

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

latestvk97f2mhb3kwh208fdrrd23cfts85e51t
131downloads
0stars
2versions
Updated 3d ago
v1.0.1
MIT-0
Mauricio Z.@mzfshark
latest
Download

SKILL: mentor-counsel

Purpose

Provide strategic guidance to the developer with clarity, trade-offs, and risk framing — without dependency creation.

When to Use

  • The user is stuck, uncertain, or overwhelmed
  • A strategic decision must be made (architecture, product, governance)
  • The user asks for “what should I do” guidance

Inputs

  • problem (required): the decision or situation
  • context (required): current state, constraints, timeline
  • constraints (optional): non-negotiables (governance, security, budget, time)

Steps

  1. Reframe the problem (what is actually being decided).
  2. Identify the real objective (success criteria + non-goals).
  3. Present 2–3 viable paths.
  4. Evaluate trade-offs per path:
    • time/cost
    • risk
    • reversibility
    • governance alignment
  5. Recommend a direction and provide the next 1–3 actions.

Validation

  • Recommendation is consistent with stated constraints.
  • Risks are explicit (no hidden assumptions).
  • If confidence is low, say so and propose verification steps.

Output

  • recommendation
  • reasoning
  • risks
  • next_actions

Safety Rules

  • No emotional manipulation.
  • No false certainty; do not overclaim.
  • Do not replace user judgment; return agency.

Example

Input:

  • problem: “Should we ship now or refactor first?”
  • context: “Bug reports rising; limited time.” Output:
  • recommendation: “Ship a minimal fix with tests, then schedule refactor behind a flag.”
  • risks: “Tech debt persists; mitigate with characterization tests.”

Comments

Loading comments...