boardroom-advisor
v1.0.0Consult a virtual board of 4 strategic advisors (Donald Miller, Seth Godin, Alex Hormozi, Daniel Priestley) on any major business decision. Two rounds of arg...
⭐ 0· 313·1 current·1 all-time
byBatsirai Chada@batsirai
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (virtual board of advisors) match the declared requirements: optional OPENROUTER_API_KEY for a stronger model, network permission to call OpenRouter, and filesystem permission to write briefs/dashboards. Nothing requested appears unrelated to running multi-round persona deliberations.
Instruction Scope
Most runtime instructions stay within scope (gather business context, run two rounds, produce files). Two items worth flagging: (1) the skill says the agent may run the board 'proactively' when it 'identifies the user is facing a major decision' — this grants broad discretion and could lead to unsolicited runs; (2) references ask users to merge OPENROUTER_API_KEY into ~/.openclaw/openclaw.json if they want the Opus model, which instructs modifying agent config. Both behaviors can be legitimate but expand the agent's autonomy and filesystem/config reach and should be reviewed by the user.
Install Mechanism
Instruction-only skill with no install spec or downloaded code. Lowest-risk install footprint; nothing is written to disk by an installer.
Credentials
Only an optional OPENROUTER_API_KEY is referenced (for running via OpenRouter/Claude Opus). No other secrets, unrelated credentials, or config paths are required. Using the key is proportionate to the stated goal of higher-quality model runs.
Persistence & Privilege
always is false and the skill does not request permanent system-wide privileges. It asks to write outputs to the working directory and optionally to update OpenClaw config per user instructions — both are reasonable for its purpose. The combination of network + filesystem + autonomous invocation increases potential impact if abused, but these are expected for this class of skill.
Assessment
This skill is coherent for running simulated advisor deliberations. Before installing: (1) decide whether you want the agent to run the board proactively—if not, ensure your agent's behavior settings prevent unsolicited runs; (2) only set OPENROUTER_API_KEY if you want to allow calls to OpenRouter (that key will enable network calls to that service); (3) expect the skill to write files (decision-brief.md, dashboard.html) to the working directory—review those outputs before sharing; (4) the personas are simulated summaries of public thinkers, not endorsements or direct representation. If you want stricter limits, avoid setting OPENROUTER_API_KEY and restrict agent autonomy so the board runs only on explicit user request.Like a lobster shell, security has layers — review code before you run it.
latestvk97f2g3rn6g5h58zq6n8zm139181z0h3
License
MIT-0
Free to use, modify, and redistribute. No attribution required.