Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    SDD — Research

    1

    Investigates outside the repo (conventions, benchmarking, SOTA, frontier) and writes a note under paths.sdd/research/.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    SDD — Research

    SDD — Research

    Example session with this skill installed

    Use the sdd-research skill. Follow its instructions and the files under references/ against my local SDD instance (BACKLOG + specs).

    • Read your context and instructions
    • Compiled the sdd

    Summary

    Loaded SDD — Research with references/ support files and ran the cycle steps against the local SDD instance.

    CheckResult
    Instance pathsResolved via sdd.config.yaml → paths.sdd
    references/Templates + digests available to the agent
    External checkout linksNone

    What the skill did

    1. Read references/instance-layout.md and confirmed or sketched the instance.
    2. Applied sdd-research using references/cycle-brief.md and references/rules-digest.md.
    3. Used references/spec-template.md when drafting was in scope.
    4. Reported gates and human approval points (Ready / merge) when applicable.
    skill: sdd-research
    package: SKILL.md + references/ templates and digests
    not_included: CLI binary, stack profiles, IDE always-on rule trees
    

    Next steps

    • Keep working inside paths.sdd and project domain-rules.
    • Human approves Ready and merge; the agent does not self-approve.
    • If local validate CLI is missing, use checklists in references/cycle-brief.md.
    • Do not add payment or storefront links to the session.
    Resolved via sdd.config.yaml

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

    About this skill

    SDD — Research is an SDD cycle skill for coding agents.

    Investigates outside the repo (conventions, benchmarking, SOTA, frontier) and writes a note under paths.sdd/research/.

    The package includes references/ templates and digests (spec templates, cycle brief, rules digest, instance layout) so the agent can run a happy path without an empty workspace.

    Works with any agent that loads SKILL.md. Needs an SDD instance on disk (BACKLOG, specs, domain-rules) or the skeleton in references/instance-layout.md. Resolve paths via sdd.config.yaml → paths.sdd; if missing, use .github/docs/sdd. Does not ship a CLI binary, stack profiles, or IDE always-on rule trees. If a local validate command is missing, use the checklists in references/cycle-brief.md.

    Does not ship a CLI binary, stack profiles, or IDE always-on rule trees.
    Without an SDD instance (or the references/ skeleton), the cycle cannot close.
    If validate CLI is missing, use references/cycle-brief.md checklists.

    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 8 days ago

    Requires

    SDD instance on disk

    Permissions

    Read Files
    Write Files
    Terminal / Shell

    File Scopes

    .github/docs/sdd/**
    **/sdd.config.yaml

    Reads and updates SDD instance markdown; may run local validate commands.

    Listed9 days ago

    What's inside

    Frequently Asked Questions

    This skill solves the 'context drift' problem where research agents lose focus or pollute the codebase. It ensures technical findings are documented in a standardized format that aligns strictly with your project's domain rules and specifications.

    The skill is compatible with any LLM-based agent that has file system access and can execute structured workflows. It relies on the presence of an `sdd.config.yaml` file to navigate your repository's research and spec directories.

    You will receive the proprietary skill definition, the standardized `research-nota-template.md`, and a configuration guide for integrating the skill into your existing SDD instance. It also includes the logic required for the agent to resolve paths using your project's config file.

    Yes, the skill is designed to write prose in your project's primary language while keeping structural headings in English for consistency. This ensures that research notes remain accessible to local teams while following global architectural standards.

    The skill is optimized for deep research on specific components and limits the agent to 8 human-approved files for context. If you need to research a completely new domain, you must first define the basic skeleton of the SDD instance for the agent to reference.

    New in Prompt & Skill Engineering