跳到正文
cccskills
免费在 GitHub 公开

cognee-switch-datasets

Use when the user wants to switch this Codex session to another Cognee dataset, see which datasets are available, or move memory capture to a different dataset. Syncs the current session first, then starts a fresh Cognee session bound to the chosen dataset.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(1)

  • SKILL.md4.1 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Switch Datasets

Move this Codex session to a different Cognee dataset. A Cognee session never spans two datasets, so the switch retires the current session (after syncing it into the graph) and starts a new one on the chosen dataset. All later saves and recalls target the new dataset; the in-context status line shows it.

Rules

  • Never guess a dataset name: list first unless the user named one.
  • Only offer datasets the listing returns — read-only ones are already filtered out.
  • Do not pass --force unless the user explicitly accepts that unsynced entries of the current session are retried at session end instead.

Instructions

1. If the user named a dataset, switch directly

python3 "${CODEX_PLUGIN_ROOT}/scripts/switch-dataset.py" "<name>" --json

2. Otherwise, list and let the user choose

python3 "${CODEX_PLUGIN_ROOT}/scripts/switch-dataset.py" --list --json

The JSON has current, datasets ([{name, id, writable, current}]), hidden_readonly (datasets you can read but not write — never offered) and filtered (true when write access was verified for every row).

Codex has no interactive picker outside plan mode, so present a numbered list of the dataset names (mark the current one) and ask the user to reply with a number or a name. If filtered is false, say that write access could not be verified. When the user answers, run the switch with the chosen name:

python3 "${CODEX_PLUGIN_ROOT}/scripts/switch-dataset.py" "<chosen name>" --json

A name that is not in the list is created for the user on switch.

3. Report the result

On success the JSON is {"switched": true, "dataset", "session_id", "previous": {"dataset", "session_id", "synced": true|false}}. Tell the user in one line which dataset and Cognee session are now active and that the previous session was synced. The status line updates on the next prompt. If "switched": false with "reason": "already_active", say the session is already on that dataset.

On failure the JSON is {"error", "code"}:

codemeaningwhat to do
2launch record not foundthe plugin did not initialise this session; run cognee-cli.sh doctor. If several Codex sessions share this directory, rerun with --session-key <host session id>
3syncing the current session failednothing was changed; show the error and ask before retrying with --force
4registering the new session failednothing was changed; the server rejected the registration — check the connection
5dataset not writablethe name is readable but owned by another principal; pick from the list

What the switch does

  1. Syncs the current session into its dataset (sync-session-to-graph.py --strict) — aborts if this fails.
  2. Ensures the target dataset exists for this principal.
  3. Registers a new Cognee session on the target dataset under a fresh connection handle, then releases the old handle (register-then-unregister, so a local agent-mode server never sees zero connections).
  4. Repoints this launch's record: every hook, the shell wrappers, the idle/exit watchers and the status line follow it.
  5. Retired sessions stay in the record's touched list; the session-end sync covers them again as a safety net.

Notes

  • A switch is not needed just to look in another dataset: a one-off graph search leaves the session where it is — ${CODEX_PLUGIN_ROOT}/scripts/cognee-search.sh "<question>" 10 --graph --dataset-id <id> (see the memory skill).

  • COGNEE_PLUGIN_DATASET only seeds the dataset at launch; a switch overrides it for the rest of the session and survives a resume.

  • Recall is scoped to the active dataset, so after a switch earlier context from the previous dataset is no longer injected — switch back to see it again.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

codebase

免费

Use when indexing a code repository into Cognee's code graph or querying it (callers, impact analysis, paths, endpoints) from Codex.

中文简介正在准备中。当前显示原文说明。

topoteretes/cognee-integrations1012026年10月20日 更新

codebase

免费

Use when indexing a code repository into Cognee's code graph or querying it (callers, impact analysis, paths, endpoints) from Antigravity.

中文简介正在准备中。当前显示原文说明。

topoteretes/cognee-integrations1012026年10月20日 更新

Explain a repository's architecture (modules, routes, storage, services), find risky hotspots, draw a Mermaid diagram of the code graph, or review how far a change reaches, from Cognee's code graph.

中文简介正在准备中。当前显示原文说明。

topoteretes/cognee-integrations1012026年10月20日 更新

Index a repository (local path or git URL) into Cognee's code graph and answer exact structural queries with no LLM calls: callers, dependencies, impact of changing a symbol, call paths, endpoints, re-index changes.

中文简介正在准备中。当前显示原文说明。

topoteretes/cognee-integrations1012026年10月20日 更新

A digest of a week or more of the user's own coding-agent sessions (default 7 days): sessions per day and project, most-edited files and the lessons Cognee learned, written up as shipped, decided and still open.

中文简介正在准备中。当前显示原文说明。

topoteretes/cognee-integrations1012026年10月20日 更新

A multi-day digest of the user's own coding-agent sessions — sessions per day and project, most-edited files, and every lesson Cognee distilled into the knowledge graph in that window. Use only when the user explicitly asks for a weekly digest, a retro, a recap or report of their recent sessions, or asks for the `cognee-digest` skill by name. Not for questions about code or the current task.

中文简介正在准备中。当前显示原文说明。

topoteretes/cognee-integrations1012026年10月20日 更新

查看 topoteretes 的全部技能

报告此技能的问题