- Home
- Skills
- Productivity
- done and dusted
Works with the AI Agents you already use
done and dusted
Automated work-journaling that captures git context and technical decisions into a centralized daily log.
$5
done and dusted
Example session with this skill installed
done-and-dusted PAY-882-fix-auth-loop
- Read your context and instructions
- Compiled the done and dusted
I have analyzed the git state for the auth-service repository. I found 3 modified files and 2 recent commits related to the authentication retry logic.
Work Journal Entry: PAY-882 fix-auth-loop
Summary
Resolved a circular dependency in the middleware that caused an infinite redirect loop when a JWT expired during an active session.
Changes
src/middleware/auth.ts: ModifiedvalidateSessionto return a 401 instead of redirecting to/loginif the request is an XHR.src/utils/token.ts: UpdatedisExpiredto handle null grace periods.- Commit
a1b2c3d: Added unit tests for the redirect boundary conditions.
Notes
Chose a 401 response for XHR to allow the frontend interceptor to handle the refresh flow. A hard redirect was breaking the client-side router state.
Next steps
- Verify the fix in the staging environment.
- Update the frontend interceptor to handle the new 401 response code.
Path: /Users/dev/daily-chore/2024-05-20/auth-service.md
Next steps
- Open the daily-chore folder to review the consolidated logs for today.
- Push the tested changes to the
feature/PAY-882branch.
Connects securely to your tools. The creator never sees your data.
About this skill
Developers often lose track of why specific technical decisions were made during a long coding session, leading to "context debt" when returning to a project weeks later. Done and Dusted solves this by automating the creation of high-context work journals based on real session data. It parses your git state and session history to write structured summaries to a central location outside your codebase, ensuring your notes are never lost in a specific branch or repo.
What it does
- Git introspection analyzes your working tree, staged changes, and recent commits to identify what actually happened.
- Context-aware summarization prioritizes the "why" behind changes, focusing on technical decisions rather than just listing modified files.
- Cross-repo tracking automatically handles multi-repository sessions by splitting entries into project-specific logs.
- Centralized storage saves all entries to a dedicated directory outside your repositories, organized by date and project name.
- Duplicate detection performs case-insensitive checks against existing logs to prevent fragmented entries for the same task.
How it works
- Trigger the entry by running the skill, optionally providing a specific task name as an argument.
- Review the evidence as the agent scans git logs, diffs, and branch names to reconstruct the work session.
- Verify the task if the agent cannot definitively identify the ticket ID or task from the branch metadata.
- Approve and save to the central documentation folder, with the option to append to an existing daily log.
Frameworks & tools
The skill requires a local environment with git installed and accessible via shell commands. It uses standard bash utilities to inspect repository state and interacts with local file systems to maintain the daily-chore directory.
Why this beats prompting it yourself
Generic prompts often result in vague summaries or hallucinated details that you have to manually correct. This skill enforces strict evidence-gathering from git and follows a specific technical template that separates committed work from conversation, ensuring your logs are accurate enough to be used as a primary source for stand-ups or PR descriptions.
Use cases
- Generating a precise "what I did today" list for daily stand-up meetings.
- Building a searchable archive of technical decisions and architectural trade-offs.
- Documenting multi-repo changes that belong to a single feature or ticket.
- Preserving work context when switching between branches or projects mid-task.
Known limitations
Requires a manual confirmation for the documentation storage path on the first run. It will not record changes in directories that are not initialized as git repositories without manual input.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Productivity.
Requires git for automatic evidence gathering. Documentation path must be confirmed on first run. Will not record secrets or environment variables.
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 buying a skill.
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 27 days ago
- Software
Git