Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Help developers build third-party tools that import, inspect, migrate, or analyze DeepChat data. Use when Codex needs to work with DeepChat provider configuration, model configuration, MCP/app settings, sessions, messages, legacy chat data, `agent.db`, `chat.db`, SQLCipher encrypted SQLite, Electron safeStorage wrapped passwords, Tauri importers, or native macOS/Windows/Linux data access.
.claude/skills/thinkinaixyz-deepchat-data-import/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -47% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 63% | 0% |
Use this skill to design or implement importers for DeepChat local data. Treat DeepChat's SQLite schema as an internal but documentable contract: inspect the current schema when precision matters, prefer read-only snapshots, and avoid writing to a live profile.
agent.db, orlegacy chat.db.
agent.db,sidecar files, encryption metadata, and backup paths.
whether the database is unencrypted, can be unlocked through Electron safeStorage, or must ask the user for the SQLite password.
settings, session, message, and legacy table relationships.
mapping DeepChat data to another app, or creating a compatibility import.
env vars, prompt text, message traces, and chat content may be sensitive.
agent.db read-write from a third-party tool. Copy agent.db,agent.db-wal, and agent.db-shm, or use SQLite backup APIs through DeepChat itself.
schema_versions, sqlite_master, and the table classesunder src/main/presenter/sqlitePresenter/tables/ before assuming column availability.
Use these repository files as the current source of truth when updating the skill or answering version-sensitive questions:
src/main/presenter/sqlitePresenter/index.tssrc/main/presenter/sqlitePresenter/connectionConfig.tssrc/main/presenter/databaseSecurityPresenter/index.tssrc/main/presenter/sqlitePresenter/tables/*.tssrc/main/presenter/agentRuntimePresenter/messageStore.tssrc/main/presenter/agentRuntimePresenter/sessionStore.tssrc/main/presenter/configPresenter/**| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 17,984 | 22,960 | +28% | 1 | 1 | 0% | 3,370 | 5,252 | +56% | 0 | 0 | — |
case-02 | fail→fail | 16,590 | 24,830 | +50% | 1 | 1 | 0% | 3,434 | 5,472 | +59% | 0 | 0 | — |
case-03 | fail→fail | 14,138 | 23,241 | +64% | 1 | 1 | 0% | 3,138 | 4,621 | +47% | 0 | 0 | — |
case-04 | pass→pass | 15,134 | 13,518 | -11% | 1 | 1 | 0% | 2,857 | 2,863 | +0% | 0 | 0 | — |
case-05 | pass→pass | 11,397 | 11,542 | +1% | 1 | 1 | 0% | 2,214 | 2,827 | +28% | 0 | 0 | — |
case-06 | pass→pass | 18,678 | 17,952 | -4% | 1 | 1 | 0% | 4,480 | 4,820 | +8% | 0 | 0 | — |
case-07 | fail→fail | 12,294 | 6,759 | -45% | 1 | 1 | 0% | 2,147 | 1,805 | -16% | 0 | 0 | — |
case-08 | fail→pass | 11,262 | 8,108 | -28% | 1 | 1 | 0% | 1,959 | 1,987 | +1% | 0 | 0 | — |
case-09 | pass→pass | 12,349 | 17,211 | +39% | 1 | 1 | 0% | 2,017 | 3,641 | +81% | 0 | 0 | — |
case-10 | pass→pass | 13,174 | 8,093 | -39% | 1 | 1 | 0% | 2,152 | 1,919 | -11% | 0 | 0 | — |
case-11 | pass→pass | 16,606 | 11,024 | -34% | 1 | 1 | 0% | 3,740 | 2,989 | -20% | 0 | 0 | — |
case-12 | fail→pass | 12,214 | 8,186 | -33% | 1 | 1 | 0% | 2,321 | 2,135 | -8% | 0 | 0 | — |
case-13 | fail→pass | 12,915 | 3,354 | -74% | 1 | 1 | 0% | 2,298 | 1,227 | -47% | 0 | 0 | — |
case-14 | fail→pass | 18,345 | 18,348 | +0% | 1 | 1 | 0% | 1,680 | 2,165 | +29% | 0 | 0 | — |
case-15 | pass→pass | 13,382 | 7,639 | -43% | 1 | 1 | 0% | 1,990 | 1,916 | -4% | 0 | 0 | — |
case-16 | fail→pass | 4,974 | 3,228 | -35% | 1 | 1 | 0% | 775 | 1,263 | +63% | 0 | 0 | — |
case-17 | fail→pass | 9,118 | 2,905 | -68% | 1 | 1 | 0% | 1,582 | 1,124 | -29% | 0 | 0 | — |
case-18 | fail→pass | 15,129 | 2,775 | -82% | 1 | 1 | 0% | 2,372 | 1,105 | -53% | 0 | 0 | — |
case-19 | fail→pass | 12,257 | 35,568 | +190% | 1 | 1 | 0% | 1,942 | 859 | -56% | 0 | 0 | — |
case-20 | fail→pass | 12,704 | 2,061 | -84% | 1 | 1 | 0% | 2,390 | 1,012 | -58% | 0 | 0 | — |
case-21 | fail→pass | 10,340 | 1,681 | -84% | 1 | 1 | 0% | 1,744 | 928 | -47% | 0 | 0 | — |
case-22 | fail→pass | 9,864 | 7,102 | -28% | 1 | 1 | 0% | 1,610 | 944 | -41% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +50 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.