CC-thread v2.1

    1

    Manage, archive, and restore project-specific conversation states and technical decisions with local Markdown indexing.

    Free

    2 usesSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    CC-thread v2.1

    Example session with this skill installed

    I've finished the auth refactor. Can you save this milestone? Call it "auth-system-v2".

    • Read your context and instructions
    • Compiled the cc-thread v2 1

    I have extracted the key decisions and milestones from our auth refactor and updated the local archive.

    Workflow Save Summary: auth-system-v2

    FieldDetails
    Workflow IDauth-refactor-2026-05-12
    Topicauth-system-v2
    StatusSuccessfully committed to local archive and project memory
    Location.cc-thread/archives/auth-refactor-2026-05-12/
    Records6 sessions (Context condensation updated)
    Key DecisionsShifted to JWT-based stateless sessions; implemented RSA256 signing.
    Next StepsIntegrate the new middleware into the production router.

    Validation Check

    • Index Integrity: Passed
    • Memory Sync: Success
    • Condensed Context: Generated (482 chars)

    Next steps

    • Run /cc-load auth-system-v2 in your next session to resume.
    • Review the context-condensed.md in your archive folder for the summary.
    • You can now safely clear the current session context.

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

    About this skill

    Context drift and session limits often force developers to repeat themselves or lose track of decision logic across long-running projects. CC-thread v2.1 solves this by providing a structured, local-first system for archiving and restoring conversation state, key decisions, and technical milestones without relying on brittle chat histories.

    What it does

    • State Restoration : Matches and reloads specific workflow contexts based on project IDs or keywords.
    • Milestone Archiving : Extracts conclusions, decisions, and next steps into a versioned local index.
    • Context Condensation : Automatically maintains a context-condensed.md for workflows exceeding five records to keep context windows lean.
    • Project Memory Sync : Dual-writes to both a readable local archive and project-scoped memory to ensure persistence across environments.
    • Startup Management : Configures opt-in startup candidates to suggest relevant workflows when a new session begins.

    How it works

    1. Save : Invoke /cc-save at a milestone. The skill extracts key points, matches them to a workflow ID, and performs an atomic dual-write to your local archive and project memory.
    2. List : Use /cc-load to view active workflows for the current project, filtered by last update and session count.
    3. Restore : Select a workflow to inject historical conclusions and unfinished items into the current session.
    4. Archive : Update metadata to hide dormant workflows from active lists without deleting the underlying data.

    Frameworks & tools

    Built for local filesystem integration using .json indexing and Markdown bodies. Supports Python 3.10+ for automated archive validation but can operate via host tools in restricted environments.

    Why this beats prompting it yourself

    Standard prompts lack a consistent storage schema, leading to fragmented notes that the LLM eventually forgets. This skill enforces atomic writes, index integrity, and a strict condensation threshold that prevents your context window from being flooded by irrelevant history.

    Use cases

    • Resuming a complex refactor after a multi-day break.
    • Syncing architectural decisions between a CLI agent and a GUI-based IDE.
    • Archiving "lessons learned" from a completed feature branch for future reference.
    • Managing multiple concurrent workstreams within the same project directory.

    Known limitations

    Does not support cross-project memory access or whole-disk backups. Context restoration is limited to 500 Unicode characters for condensed files. Requires local file write permissions.

    Local file access required. Restoration body limited to 500 characters when condensed. No cross-project memory retrieval. Python 3.10+ needed for automated validation script.

    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

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    It uses a dual-write system that stores data in both a human-readable local Markdown archive and a project-scoped memory index, ensuring your decisions remain accessible even if you switch between different IDEs or CLI agents.

    The skill requires a host environment capable of local file system operations (Python 3.10+ recommended) and works best with agents that support tool-use for reading and writing Markdown and JSON files.

    When a workflow exceeds five records, the skill triggers a 'Context Condensation' process that summarizes the history into a lean file, preventing your LLM's context window from becoming cluttered with redundant data.

    You receive the CC-thread v2.1 skill package, including the core logic for state restoration and milestone archiving, local storage schema templates, and documentation for startup management configuration.

    Version 2.1 introduces the automated 'context-condensed.md' feature and the dual-write sync mechanism, significantly improving upon the manual archiving methods found in previous iterations.

    New in Workflow Automation