协作头脑风暴
brainstorming
在实现前澄清需求、探索设计选择,并形成清晰的开发方向。
record-and-replay【开发与工程】Use when the user wants Codex to record a Linux desktop or browser workflow and turn it into a reusable skill. Requires the Record & Replay event-stream MCP server.
独立 GitHub 快照 · 自动分类 · 以原作者文档为准
Record the user's Linux workflow, inspect the event stream, and turn the repeatable intent into a reusable Codex skill. This is the same bundled Record & Replay product shell backed by a Linux-native event-stream server, not a raw pointer or keyboard macro recorder.
The bundled plugin launches ./bin/SkyLinuxComputerUseClient event-stream mcp.
That helper is the Linux equivalent of the macOS Sky Computer Use event-stream
client, implemented by the Rust codex-record-replay-linux backend.
doctor when readiness is uncertain. If Computer Use cannot observe the
desktop, explain the blocker and offer the diagnostics/setup path before
recording.skysight_start when recent activity context should keep accumulating
before or during the demo, or skysight_snapshot when you only need a
point-in-time local activity summary. Use skysight_pause and
skysight_resume to stop or continue Chronicle-compatible resources
without losing the active session, and use skysight_status to find the
resource paths and local OCR availability. Respect
skysight_list_exclusions and update exclusions before recording sensitive
apps or domains. Treat OCR as local screen evidence metadata; do not copy raw
OCR text into durable skill drafts unless it is necessary for the reusable
workflow and safe to retain.event_stream_start with a short goal when matching the upstream
Record & Replay flow, or start when you need Linux-specific options. The
Linux app should show the active Record & Replay recording HUD while the
shared runtime status is active. Native Linux audio evidence is opt-in and
requires both include_audio: true and an affirmative
CODEX_RECORD_REPLAY_AUDIO setting; normal workflow recording should rely
on transcript speech_context instead.
Tell the user that recording is active, that they should perform the workflow
normally, and that they can say done when finished.desktop_snapshot at meaningful app/window
changes, such as after the user opens Chrome or arrives on a target site.
Call mark only for meaningful intent boundaries that will help the future
skill, such as "source page opened" or "finished selecting rows".speech_context with the transcript. Treat the speech as user
intent/context, not as audio to replay or Chronicle-compatible resources.
Do not hijack the composer dictation UI as the recording architecture.browser_trace when browser/CDP trace evidence
is available. Treat the trace as semantic evidence for drafting the skill,
not as a click/coordinate replay script.event_stream_status or status if you need to confirm which bundle is
active. When the user says they are done, asks to stop, or the HUD sends
"I'm done recording.", call event_stream_stop or stop if the bundle is
still active.event_stream_cancel or cancel with discarded: true and treat the bundle
as canceled evidence only.validate_bundle, then draft_skill_prompt.SKILL.md. Prefer stable app names, URLs, semantic UI labels, and data
shape descriptions over literal coordinates.inspect_skill before import. Call import_skill only after the user
approves the generated skill.