ctx-flow: Project Memory for Coding Agents

    1

    Keep project goals, decisions, and domain notes concise, then load the context each coding task needs.

    Free

    2 usesSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    ctx-flow: Project Memory for Coding Agents

    Example session with this skill installed

    Illustrative example using sample project notes:

    Use ctx-flow to prepare an auth handoff. The project goal is to ship passkey login. The constraints are to keep password login available and avoid storing credentials. We decided to add passkeys incrementally. The auth domain note says WebAuthn is the chosen approach. The UI domain has unrelated dashboard layout notes. Load only the relevant context and summarize what is known and what remains open.

    • Read your context and instructions
    • Compiled the ctx-flow

    Auth handoff

    This illustrative example shows how ctx-flow organizes the supplied sample facts. It does not claim that an implementation has been completed.

    Context selected

    FileRelevant information
    context/goals.mdShip passkey login.
    context/constraints.mdKeep password login available; do not store credentials in project memory.
    context/decisions.mdIntroduce passkeys incrementally.
    context/domains/auth.mdWebAuthn is the chosen approach.

    The unrelated dashboard layout notes in context/domains/ui.md were not needed for this task.

    Verified scope

    The supplied notes establish the desired feature, constraints, and chosen approach. They do not establish which endpoints, browser flows, or tests already exist.

    Open questions

    • Which registration and authentication flows are already implemented?
    • How should existing users enroll a passkey?
    • What recovery behavior is required when a passkey is unavailable?

    Next step

    Inspect the current authentication code and tests, then update the auth notes with verified implementation facts. Preserve the password-login constraint and record decisions only after they are settled. No API key or background MCP server is needed for this local Markdown workflow.

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

    About this skill

    Keep useful project context across coding sessions

    ctx-flow helps your coding agent maintain concise Markdown notes for project goals, constraints, decisions, and domain-specific facts. It reads the core notes and selects relevant domains for the current task instead of loading every note.

    What is included

    • A portable SKILL.md workflow for maintaining project memory.
    • Guidance for updating verified facts, resolving stale notes, and preparing handoff summaries.
    • An optional MCP setup reference for keyword routing, persistent conversation history, and Anthropic-powered extraction and summarization.

    How to use it

    Ask your compatible agent to use ctx-flow in your project. It inspects existing context files, loads relevant notes, and records durable facts supported by your instructions or verified project changes.

    Requirements and limits

    The core workflow needs local file-reading and file-writing tools. It does not require an API key. The optional MCP server is a separate installation; its AI helpers require an Anthropic API key and may incur API charges. Installing this skill does not register the server, intercept every prompt, or compact the host agent's conversation automatically. Summaries are lossy and no fixed token-saving percentage is guaranteed.

    Free and open source

    Created by the maintainer of ctx-flow and offered free under the MIT license. The skill and related source are public at https://github.com/bansalbhunesh/ctx-flow/tree/fix/skill-packaging-and-context-safety/skills/ctx-flow . The optional server has its own setup instructions; hosted services and paid support are not included.

    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

    Permissions

    Read Files
    Write Files

    File Scopes

    context/**

    Core mode reads and writes only the selected project context directory. No account connection or network access is required. Optional MCP installation uses a shell; enabling Anthropic helpers uses an environment-provided API key and sends relevant messages, notes, and history to Anthropic. Do not store credentials or private correspondence in context notes.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    Keep project goals, decisions, and domain notes concise, then load the context each coding task needs.

    This skill uses the open SKILL.md standard. Use it through an MCP-compatible agent connected to Agensi, and check the compatibility information on this page for confirmed agents and required capabilities.

    Connect your AI agent to Agensi through MCP, then ask it to use ctx-flow: Project Memory for Coding Agents on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.

    New in Prompt & Skill Engineering