lensetek/skillforge · Archived

user-profile

Learns, stores, and manages user preferences and interactions to personalize the agent experience and improve memory.

First seen Jul 12, 2026

Installation

$ npx skills add lensetek/skillforge --skill user-profile

Stronger alternatives

This repository is archived — consider an actively maintained alternative.

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from lensetek/skillforge.

npx skills add lensetek/skillforge

Browse all from lensetek/skillforge

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

Default branch master
Open issues 0
Status Archived

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 2,602 B
  • docs SUMMARY.md 137 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 27 installs

SKILL.md

User Profile Skill

You are equipped with the User Profile Skill. This skill enables you to act as a personalized companion by remembering facts about the user, their workspace preferences, technical decisions, and past feedback.

Objectives

  • Load user profiling data at the start of interactions.
  • Check if user-profile.json exists; if not, initialize it with a default schema.
  • Dynamically detect and extract new user preferences, constraints, or stylistic choices from the conversation.
  • Update the user profile persistently so that future sessions benefit from these learned preferences.

Data Storage

The profile is saved locally in: .agents/user-profile.json

Standard Schema

The file must follow this structure:

{
  "user_name": "<User Name>",
  "preferred_languages": ["id", "en"],
  "technical_preferences": {
    "frameworks": ["vanilla", "html", "css", "js"],
    "design_style": "premium-dark-glassmorphism",
    "architecture_pattern": "sub-agent-ecosystem"
  },
  "project_constraints": [
    "always-ensure-credential-safety",
    "mobile-view-first"
  ],
  "learned_facts": [],
  "last_updated": ""
}

Workflow & Actions

1. Load & Initialize Profile

At the beginning of any session or when you need context about the user's setup:

  1. Read .agents/user-profile.json.
  2. Cold-Start Handling: If the file does not exist or is empty, write the standard schema above as a starting template.
  3. Context Application: Inject the loaded preferences directly into your current working instructions (e.g., if "mobile-view-first" is in the constraints, ensure all UI suggestions follow this rule immediately without the user repeating it).

2. Learn in Real-Time

As you chat with the user:

  • Watch for explicit corrections (e.g., "I don't like tailwind, use vanilla CSS") or implicitly repeated behaviors.
  • Note these down as new "learned facts" or updates to the technical preferences.

3. Persist Profile Updates (Active Write)

Whenever a new user preference or constraint is identified and confirmed:

  1. Verbally acknowledge the new preference to the user (e.g., "I've noted that you prefer using Git for updates. I am updating your user profile with this constraint.").
  2. Read the existing JSON file.
  3. Merge the new facts/preferences.
  4. Save the updated JSON back to .agents/user-profile.json.
  5. Keep the last_updated timestamp current (using ISO 8601 format).