Coding Agent Personalizer

Coding Agent Personalizer by ITMO turns recent Git activity into reviewable suggestions for the instruction files and reusable skills used by coding agents.

The extension collects local repository signals, sends bounded summaries to the Coding Agent Personalizer service, and keeps you in control before it writes any rule or skill file.

Requirements

• Visual Studio Code 1.118.0 or later
• A Git repository opened as the first folder in the VS Code workspace

Quick Start

1. Open a Git repository in VS Code.
2. Work normally: edit files, run tests, and make commits as usual.
3. After you stop changing files for the idle window, Coding Agent Personalizer closes the current coding episode and evaluates whether enough changed to refresh your profile.
4. When the extension finds useful rule suggestions, it shows a notification and a status bar item.
5. Open the review view from the notification, status bar item, or Coding Agent Personalizer Activity Bar view.
6. Accept, edit, or reject each suggestion.
7. Preview the proposed rule-file change.
8. Apply the accepted suggestions only when the preview looks right.

Nothing is written to your repository until you apply accepted or edited suggestions. Pending and rejected suggestions are never applied.

How It Works

Coding Agent Personalizer runs as a review loop around your normal work:

1. It watches workspace file changes and groups them into coding episodes.
2. When you stop editing for the configured idle window, it summarizes the episode using changed paths, line counts, duration, top directories, and top file extensions.
3. If the episode passes the configured duration, changed-line, and cooldown thresholds, the extension refreshes your profile from recent Git activity.
4. Coding Agent Personalizer compares the updated profile with the existing agent rule file and proposes short, reviewable improvements.
5. You review every suggestion before anything is written.
6. Accepted and edited suggestions can be previewed as a VS Code diff, then applied to the selected rule file.

The Coding Agent Personalizer Activity Bar view shows the latest repository, identity, developer profile, rule review, and skill candidate state.

Automatic Personalization

Automatic personalization is enabled by default. The extension groups workspace file changes into coding episodes and waits for an idle period before it asks whether the work was meaningful enough to analyze.

A coding episode can trigger new rule suggestions only when it passes all of these checks:

• the episode lasted at least codingAgentPersonalizer.automation.minDurationSeconds
• the working tree changed at least codingAgentPersonalizer.automation.minChangedLines
• the cooldown window in codingAgentPersonalizer.automation.cooldownMinutes has passed

The default idle window is codingAgentPersonalizer.automation.episodeIdleSeconds, which is 60 seconds.

Disable automatic personalization with:

{
  "codingAgentPersonalizer.automation.enabled": false
}

Rule Suggestions

With the default auto target, the extension uses an existing supported rule file when one is present. If .cursor/rules exists without a rule file, it creates .cursor/rules/agent-profile.mdc; otherwise, it creates AGENTS.md.

Supported targets include:

• AGENTS.md and agents.md
• .github/copilot-instructions.md
• .cursor/rules/agent-profile.mdc and existing Cursor rule files
• CLAUDE.md
• .codex/AGENTS.md

The preview step opens a VS Code diff without writing to the repository. Applying approved suggestions writes either a managed Personalized Agent Rules section or a targeted edit supplied by Coding Agent Personalizer.

Skill Candidates

Coding Agent Personalizer can identify repeated, task-specific workflows from classified commit history and return them as generated skill candidates. Skill files are never exported without your approval.

Skill review is commit-driven. The extension watches for new commits and checks whether any commit crosses codingAgentPersonalizer.skills.minCommitChangedLines. Large enough commits can trigger a skill-candidate scan. When generated candidates are ready, you can review them from the notification or the Coding Agent Personalizer Activity Bar view.

To review and export candidates:

1. Open the skill review view from the notification or Coding Agent Personalizer view.
2. Accept or reject each candidate.
3. Select Export Accepted.

The portable .agents/skills/<skill-name>/SKILL.md target is always included. The extension can also detect existing tool-specific configuration and export to compatible skill directories for Claude, Cursor, GitHub, Gemini, Windsurf, Cline, Roo, and OpenCode.

Existing manually authored skills are skipped by default. Enable codingAgentPersonalizer.skills.overwriteExisting only when you intend to allow replacement of those files.

Manual Controls

The automatic workflow covers the normal path. The Command Palette also exposes manual controls when you want to scan immediately, open the profile output, generate suggestions on demand, review suggestions, preview or apply accepted rules, review skill candidates, record a prompt note, or reset this workspace's local cache.

Settings

Data and Privacy

Before using the extension, confirm that Coding Agent Personalizer is approved for your repository data.

The extension sends the Coding Agent Personalizer service:

• Recent Git commit metadata, changed paths, and line counts
• Repository remotes, branches, HEAD information, and locally available pull request references
• Stable user identity evidence derived from VS Code authentication, Git configuration, repository remotes, or a generated local identifier
• Existing rule-file text when requesting rule suggestions
• Coding episode summaries and optional notes you explicitly record
• Local profile cache data needed to continue the review workflow

The current extension does not send raw diff snippets or automatically read prompt history from other AI tools.

Workspace review state is stored in a hidden .profile cache inside VS Code-managed workspace storage, not in the repository. Reset Local Profile Data clears that local cache only; it does not delete server-side history.

Troubleshooting

An action says to open a workspace folder

Open a Git repository as the first folder in the VS Code workspace.

No rule suggestions are available

Let the automatic workflow observe meaningful repository activity first, or start a manual scan from the Coding Agent Personalizer view. Suggestions depend on the available Git history and the selected developer profile.

No skill candidates are available

Skill candidates require repeated workflow evidence in commit history. Ordinary coding episodes do not request new skill candidates.

The extension did not update the file I expected

Set codingAgentPersonalizer.rules.preferredFile to a specific supported target, or review the detected target in the rule suggestion view before applying.