Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    The Council

    3

    Convene an independent multi-model council to stress-test decisions and eliminate AI sycophancy.

    Free

    2 usesSecurity scanned
    The Council

    The Council

    Example session with this skill installed

    council deep "I'm planning to migrate our entire Postgres RDS to a managed ClickHouse instance for better query performance. Is this a mistake or should I go ahead?"

    • Read your context and instructions
    • Compiled the the council

    The council has concluded its deep-mode deliberation on the proposed database migration. Three internal seats and two outside seats (GPT-4o, Gemini 1.5 Pro) were convened.

    Council Verdict: AGNOSTIC / PROCEED WITH CAUTION

    The council found that while your premise regarding query performance is likely true for analytical workloads, the migration poses a high risk to transactional integrity which was not addressed in your prompt.

    SeatVoteConfidencePrimary Reasoning
    Claude OpusOpposeHighClickHouse lacks full ACID compliance for your current transaction patterns.
    Claude SonnetSupportMediumOLAP performance gains (est. 10x-50x) outweigh the engineering overhead.
    Claude FableOpposeHighFound 4 core services relying on Postgres-specific JSONB functions.
    GPT-4oSupportLowScaling RDS is more expensive than a basic ClickHouse cluster.
    Gemini 1.5 ProOpposeMediumLatency for single-row lookups will regress significantly.

    Verified Findings

    Claim: "ClickHouse is faster for our queries." ->

    Verified. Synthetic benchmarks on your events table show 40x speedup.

    Claim: "Migration is straightforward." ->

    Refuted. Grep found 12 stored procedures that require manual rewriting.

    Minority Report (Sonnet)

    The migration is viable if the architecture is split. Move the analytics schema to ClickHouse but retain Postgres for orders and users to maintain transactional safety.

    Next steps

    1. Review the references/sql-incompatibility.log generated during the audit.
    2. Perform a latency test on single-row SELECT queries to simulate dashboard lookups.
    3. Consider a dual-write strategy rather than a full cutover.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Developers often struggle with AI "yes-man" bias, where models reflexively agree with a user's stated preference or plan. This skill forces objective scrutiny by convening a multi-model council that operates through a hardened workflow to ensure honesty.

    What it does

    • Multi-model voting triggers independent ballots from Claude Opus, Sonnet, and Fable to eliminate single-model blind spots.
    • Blind framing rewrites user requests into neutral, symmetric questions so voters cannot detect the user's desired outcome.
    • Premise verification uses automated tools to audit load-bearing claims before a verdict is reached.
    • Red teaming automatically triggers a dissent round if the council unanimously agrees with the user on high-stakes decisions.
    • Outside seats allows optional integration of GPT or Gemini models to provide vendor-diverse perspectives.

    How it works

    1. Intake involves parsing your question, identifying your stated stance, and gathering necessary repository context.
    2. Neutralization occurs via a framer agent that shuffles options and removes evaluative language to prevent anchoring.
    3. Balloting runs independent seats with verification tool access to check facts against the codebase or web.
    4. Adjudication by a non-voting chair rules on evidence and compiles a minority report for any dissenting views.

    Frameworks & tools

    This skill utilizes the Claude Code Workflow tool, jq for result parsing, and local shell scripts for optional third-party model integration.

    Why this beats prompting it yourself

    Standard prompting fails to hide user intent from the model, leading to sycophancy. This skill uses a programmatic workflow to physically separate the "framer" from the "voters," making it impossible for the judging models to simply tell you what you want to hear.

    Use cases

    • Architecture reviews to determine if a proposed migration or refactor is genuinely the best path.
    • Sanity checks for complex debugging theories where you might be chasing a red herring.
    • Strategic decisions where you need a "devil's advocate" to poke holes in a business or technical plan.
    • Security assessments to red-team a proposed auth flow or permission model.

    Known limitations

    Requires Claude Code or a compatible Workflow runner. Outside seats (GPT/Gemini) require local CLI configuration and existing API billing. Deep mode may consume significant token quotas.

    Requires Claude Code Workflow tool. Outside models need local CLI setup. Not for simple lookups; designed for judgment calls. Workflow must be installed to ~/.claude/workflows.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    2 uses

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Requires

    Python

    Permissions

    Terminal / Shell
    Read Files
    Browser
    Network Access
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    The skill solves the "yes-man" bias (AI sycophancy) by using a blind framing process that hides your preferred outcome from the judging models, forcing them to evaluate your proposal objectively rather than just agreeing with you.

    Council is designed for the Claude Code environment and compatible workflow runners; while it defaults to Claude models, you can integrate GPT-4 or Gemini via local CLI configurations for a more diverse range of perspectives.

    You will receive the complete Council workflow package, including the neutralizer and adjudicator logic, local shell scripts for third-party model integration, and the 'jq' parsing configurations required to run the multi-model ballots.

    Setup involves importing the workflow into your Claude Code environment and, if you choose to use 'Outside Seats,' ensuring your local environment has the necessary API keys and CLIs configured for non-Anthropic models.

    The skill includes a 'Red Teaming' feature that automatically triggers a dissent round if the council is in unanimous agreement on high-stakes decisions, ensuring a devil's advocate perspective is always considered.

    Yes, the skill is updated to maintain compatibility with new versions of the Claude Code tool and to refine the 'Premise Verification' logic as new automated auditing tools become available.

    New in Code Quality & Review