What does this skill do, and when should you use it?
This skill assists developers in designing importers for DeepChat's local data. It covers locating the data (agent.db, chat.db), understanding the SQLite schema, handling encryption (SQLCipher, Electron safeStorage), and safely extracting data without modifying the active database. The skill provides reference files for data locations, SQLite access, schema details, and import recipes. It emphasizes security: obtain user consent, use read-only snapshots, and redact secrets by default.
Reads local DeepChat data files (agent.db, chat.db). Inspects SQLite schema for providers, settings, sessions, messages. Uses Electron safeStorage to decrypt passwords. Provides guidance for data migration and analysis. Does not write to active databases.
- A developer migrating chat history from DeepChat to another tool, using the skill to map data.
- A developer analyzing DeepChat usage patterns for debugging or product insights.
- A team building a backup or sync service for DeepChat data.
- An auditor needing to export DeepChat data for compliance checks.
- A security researcher investigating DeepChat's storage security.
How do you install this skill?
- The skill involves reading sensitive data; ensure explicit user consent and default redaction to avoid leaking keys and chat content.
- Encrypted DB unlock relies on Electron safeStorage; cross-app/machine decryption is unstable; prepare password fallback.
- The skill depends on DeepChat-specific schema which may change; verify against current source code.
- Mainland network may block access to overseas resources; assess network reachability for runtime.
- Shell / CLI
- Local filesystem
SQLCipherElectron safeStorage
The skill is part of the DeepChat repo (ThinkInAIXYZ/deepchat) at .agents/skills/deepchat-data-import/. To install independently, clone or download the skill folder from [GitHub](https://github.com/ThinkInAIXYZ/deepchat) and copy it to your agent's skills directory (e.g., .agents/skills/).
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/ThinkInAIXYZ/deepchat.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/.agents/skills/deepchat-data-import" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Import my DeepChat conversations into XYZ
- Inspect this agent.db file
After installation, assign a task like 'Import my DeepChat conversations into XYZ' or 'Inspect this agent.db file'. The skill will guide the agent through reference files for data locations, database access, schema, and import recipes. It requires user consent and uses WAL-aware snapshots or backups.
What are this skill's strengths and limitations?
- Provides comprehensive references: data locations, SQLite access, schema, and import recipes.
- Prioritizes privacy and safety: explicit user consent, read-only snapshots, no live database writes, automatic secret redaction.
- Addresses encryption complexities: guides handling SQLCipher and Electron safeStorage.
- Includes source files for version-sensitive queries.
- Only covers DeepChat data import, not other applications.
- Reference files are not included in the prompt; requires fetching from the repo.
- No test suite or validation tools mentioned, potentially reducing reliability.
- Only verified on the codebase, not cross-platform tested.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| DeepChat Data Import Skill this page | 62 · Recommended | ★ 6.4k | 3d ago | Apache-2.0 |
| Desktop App Control with Computer Use | 61 · Recommended | ★ 6.4k | 3d ago | Apache-2.0 |
| Weaviate Database Operations Skill | 50 · Use with care | ★ 105 | 5d ago | BSD-3-Clause |
| Engram Persistent Memory Protocol | 68 · Recommended | ★ 7.1k | 4d ago | MIT |
| Jetpack Navigation 3 Guide and Recipes | 59 · Recommended | ★ 7.7k | 16d ago | Apache-2.0 |
This skill is specific to DeepChat; no direct alternatives, as it deals with a unique application's data format.
How did FollowSkills review this skill?
The skill explicitly requires user consent, read-only snapshots, avoiding writes to live profile, parameterized password APIs, default redaction, and emphasizes sensitive data handling. No code-level least privilege proof, but risks are controlled; deductions for lack of in-depth permission verification and rollback details.
The skill provides decision trees and error handling suggestions (e.g., SQLITE_NOTADB, missing WAL), but lacks tests or CI coverage, static review cannot verify execution. Deductions for no executed tests and possibly incomplete error handling.
Audience and scenarios clear, supports Chinese users, but mainland network reachability not proven, and platform features like Electron safeStorage may be limited. Deductions for insufficient environment fit evidence.
Documentation well-layered with references and examples, but lacks versioning, changelog, and maintenance responsibility info. Deductions for missing versioning and governance.
Provides concrete steps and output format, but no actual output examples or comparative benchmarks; static review cannot verify direct usability. Deductions for lack of executed verification and comparative benefit.
References repository source files but no tests or reproducible experiments; static review cannot verify. Deductions for lack of executable verification material.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →