Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Decision Log

    1

    A strict ADR generator that logs single architecture decisions from source dumps without fluff or hallucinations.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Decision Log

    Decision Log

    Example session with this skill installed

    Log the decision to use Redis for the session store. Context: we need faster reads than Postgres for sessions. Options: Redis vs Memcached. Picked Redis because of persistence. Date: 2023-10-12. Role: Lead Dev.

    • Read your context and instructions
    • Compiled the decision log

    I have drafted the Architecture Decision Record (ADR) based on your notes. The log has been prepared in docs/decisions.md with status ready.

    Decision Log: Redis for Session Store

    Header

    • Decision: Redis for Session Store
    • Date: 2023-10-12
    • Decided by: Lead Dev
    • Status: ready
    • Write Mode: draft

    Context
    The team requires a session storage solution with lower latency than the current PostgreSQL implementation to handle increased concurrent traffic.

    Options

    IDNameSummarySource
    O1RedisKey-value store with persistence capabilitiesbuyer paste
    O2MemcachedHigh-performance distributed memory object caching systembuyer paste

    Decision

    • Picked: O1

    Statement: Use Redis for the session store to achieve faster reads than Postgres while maintaining session persistence.

    • Source: buyer paste

    Consequences

    • Change: Session data will migrate from Postgres to a dedicated Redis instance.
    • Source: derived

    Rejected

    • O2 (Memcached)

    Diff

    • New ADR appended to docs/decisions.md.

    Gaps

    • (none)

    Residual

    • mutated: no
    • invented: no

    Next steps

    1. Review the generated docs/decisions.md file.
    2. If the draft is accurate, run with write_mode=apply to update the project wiki.
    3. Notify the infrastructure team regarding the new Redis dependency.

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

    About this skill

    The problem

    Engineering and ops teams often lose the "why" behind technical choices, leading to repetitive debates or accidental reversals. Traditional ADRs frequently suffer from "architecture astronautics," becoming bloated with diagrams and non-functional requirements that obscure the actual decision.

    What it does

    • Extracts specific technical context and candidate options from raw developer dumps or notes.
    • Enforces a strict requirement for at least two sourced alternatives to ensure a real choice was made.
    • Generates dated, role-attributed ADRs in Markdown format without inventing "do-nothing" filler or dates.
    • Identifies gaps like contradictions in the source material or attempts to include out-of-scope architectural diagrams.
    • Manages a local decisions.md log, appending new records while maintaining a clear audit trail.

    Frameworks & tools

    This skill is framework-agnostic. It outputs standard GitHub-flavored Markdown for use in docs/ folders, Git repositories, or engineering wikis.

    Why this beats prompting it yourself

    General-purpose prompts often hallucinate "standard" options like "do nothing" or invent dates and stakeholders to look complete. This skill uses a strict binding and freezing procedure that refuses to invent data, ensuring your logs reflect reality rather than LLM assumptions.

    Use cases

    • Logging the selection of a specific database or cloud provider from raw meeting notes.
    • Documenting infrastructure changes where exactly one choice was made between two specific tools.
    • Capturing the rationale for a breaking API change before the context is lost.

    Known limitations

    Requires an ISO date, a specific decision title, and a defined role. It will block execution if the source material contains only one option or contradictory winners.

    Requires ISO date, title, and role at runtime. Blocks if <2 options found or if winner is contradicted. Does not support multi-decision files, C4 diagrams, or meeting transcript extraction.

    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

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 2 months ago

    Permissions

    Read Files
    Write Files
    Listed2 months ago

    What's inside

    Frequently Asked Questions

    The skill requires a specific decision title, an ISO-formatted date, a defined stakeholder role, and at least two distinct technical alternatives extracted from your source notes. If these elements are missing or contradictory, the skill will flag the gaps rather than hallucinating filler data.

    This skill is framework-agnostic and generates standard GitHub-flavored Markdown, making it compatible with any Git repository, internal engineering wiki, or documentation folder.

    Unlike general prompts that often invent 'placeholder' dates or generic 'do-nothing' options, this skill uses a strict binding procedure that refuses to generate content not found in your source material, ensuring a 100% accurate audit trail.

    You will receive the compiled AI agent skill, including the specialized logic for context extraction and ADR formatting, along with a guide for integrating it into your existing development workflow.

    Yes, the skill is designed to manage and append to a local 'decisions.md' file, allowing you to maintain a chronological and consistent record of all architectural choices over the life of a project.

    New in Agents & Orchestration