- Home
- Skills
- Agents & Orchestration
- road map journal
Works with the AI Agents you already use
road map journal
Keep roadmaps and work journals synced with your codebase through evidence-based status tracking and decision logging.
$5
road map journal
Example session with this skill installed
Update the roadmap. I just finished the auth-middleware unit tests and they are passing, but the integration tests are still pending.
- Read your context and instructions
- Compiled the road map journal
I have updated the project roadmap and work journal to reflect the current state of the authentication middleware. I verified the unit test completion against the local test reports and flagged the remaining integration work.
Development Roadmap Update
| Item ID | Task Description | Status | Evidence / Notes |
|---|---|---|---|
| AUTH-01 | Auth Middleware Implementation | PARTIAL | Unit tests passed; Integration tests pending. |
| AUTH-02 | Session Store Integration | OPEN | Blocked by AUTH-01. |
| CORE-05 | Environment Config Refactor | DONE | Verified in commit a7b2d3f. |
Work Journal Entry: 2023-10-27
- Action: Updated status for
AUTH-01. - Findings: Unit tests for
auth-middleware.tsare 100% passing (12/12).
Decisions: Maintaining PARTIAL status instead of DONE because integration coverage is required by the project acceptance criteria.
- Next Steps: Proceed to integration testing once the mock database is configured.
Next steps
- Review the integration test suite requirements in
docs/testing.md. - Confirm if the mock database container is ready for the
AUTH-01integration phase. - Signal when you are ready to move
AUTH-02to IN PROGRESS.
Connects securely to your tools. The creator never sees your data.
About this skill
Developers often struggle with fragmented project history and stale roadmaps that don't reflect the actual codebase state. This skill synchronizes your development journal and planning documents with the reality of your version control and implementation progress.
What it does
- State Synchronization reads your version control and project root to align written plans with actual code changes.
- Journal Maintenance records decisions, findings, and evidence in a traceable history without overwriting previous context.
- Status Reconciliation updates item statuses (OPEN, IN PROGRESS, BLOCKED, DONE) based on concrete evidence rather than intent.
- Queue Management preserves the order of planned work while marking superseded or cancelled tasks with their justifications.
- Multi-Agent Coordination checks for dirty paths and overlapping edits before writing to files in shared environments.
How it works
- Discover: The skill scans your project root for existing roadmaps, handoff notes, and journal entries.
- Select Mode: It enters Read-only, Init, Adopt, Resume, or Update mode based on your specific prompt.
- Analyze: It compares the recorded plan against the current implementation and version-control state.
- Propose: For significant changes or new structures, it proposes edits and waits for your consent.
- Execute: It applies narrow, scoped edits to your documentation, preserving unrelated work and historical context.
Frameworks & tools
This skill is agnostic to specific documentation formats or folder structures. It works with standard version control systems and any project layout using Markdown or text-based journals and roadmaps.
Why this beats prompting it yourself
Generic prompts often result in the AI hallucinating progress or deleting historical context when updating a list. This skill enforces strict evidence-based status updates and preserves the "why" behind decisions, ensuring your roadmap remains a reliable source of truth for the entire team.
Use cases
- Resuming work on a complex feature by reading the last recorded state and acceptance criteria.
- Auditing a roadmap to mark completed items as DONE based on actual code evidence.
- Initializing a structured work journal for a new repository without inventing fake priorities.
- Updating a shared development plan after another agent or developer has made commits.
Known limitations
Does not support automatic publishing or committing to version control without explicit authorization. Requires accessible text-based planning documents to function in Adopt or Resume modes.
Requires existing text-based plan or journal for Resume/Adopt modes. Cannot commit or publish changes to version control without explicit user authorization.
How to use
- 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
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 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 12 days ago