refactor(cli): --limit/--all-pages, Go 1.26, internal/agent → aiclient
Cross-cutting cleanup that lands alongside the new feature surface: - `--limit / -L` and `--all-pages` on every list command. Default --limit 30 (gh-parity); --all-pages drains every server page client-side, capped by --limit. Closes the audit finding that the old "1000 max per call" implicit cap was undiscoverable. - `auth token` emits a TTY-only stderr advisory when stdout is a terminal (the credential just got displayed in scrollback) plus an api-key-mode rotation hint. - Comment + doc discipline pass: drop external project name references from in-code comments (we reference them in design notes, not inline). - Bump `go` directive to 1.26.0 and CI matrix to 1.26.x to align with the main module's go.mod. - Rename `cli/internal/agent` → `cli/internal/aiclient` to disambiguate from the new `cli/cmd/agent` resource subtree. The package handles AI coding-agent env detection + per-command --help annotations; the new name reflects that more precisely.
This commit is contained in:
+10
-3
@@ -27,7 +27,7 @@ import (
|
||||
|
||||
"github.com/spf13/cobra"
|
||||
|
||||
"github.com/Tencent/WeKnora/cli/internal/agent"
|
||||
"github.com/Tencent/WeKnora/cli/internal/aiclient"
|
||||
"github.com/Tencent/WeKnora/cli/internal/cmdutil"
|
||||
"github.com/Tencent/WeKnora/cli/internal/format"
|
||||
"github.com/Tencent/WeKnora/cli/internal/iostreams"
|
||||
@@ -38,7 +38,8 @@ import (
|
||||
// chatFields enumerates the fields surfaced for `--json` discovery on `chat`.
|
||||
// Mirrors the chatData struct json tags.
|
||||
var chatFields = []string{
|
||||
"answer", "references", "session_id", "assistant_message_id", "kb_id", "query",
|
||||
"answer", "references", "thinking",
|
||||
"session_id", "assistant_message_id", "kb_id", "query",
|
||||
}
|
||||
|
||||
type Options struct {
|
||||
@@ -62,6 +63,11 @@ type chatService interface {
|
||||
type chatData struct {
|
||||
Answer string `json:"answer"`
|
||||
References []*sdk.SearchResult `json:"references"`
|
||||
// Thinking holds the reasoning / reflection text emitted by reasoning
|
||||
// models (GPT-5, Claude extended thinking) via response_type=thinking
|
||||
// frames. Omitted when empty (non-reasoning model or model didn't
|
||||
// surface reasoning for this query).
|
||||
Thinking string `json:"thinking,omitempty"`
|
||||
SessionID string `json:"session_id"`
|
||||
AssistantMessageID string `json:"assistant_message_id,omitempty"`
|
||||
KBID string `json:"kb_id"`
|
||||
@@ -110,7 +116,7 @@ Modes:
|
||||
cmd.Flags().StringVar(&opts.SessionID, "session", "", "Continue an existing chat session (skip auto-create)")
|
||||
cmd.Flags().BoolVar(&opts.NoStream, "no-stream", false, "Buffer the full answer before printing (forces accumulate mode)")
|
||||
cmdutil.AddJSONFlags(cmd, chatFields)
|
||||
agent.SetAgentHelp(cmd, "Streams an LLM answer over SSE. Agent / non-TTY callers should pass --json so the full {answer, references, session_id, assistant_message_id} envelope is emitted at completion (no partial chunks). Pass --session to thread follow-ups. Errors: server.session_create_failed when auto-create fails; local.sse_stream_aborted on mid-stream disconnect.")
|
||||
aiclient.SetAgentHelp(cmd, "Streams an LLM answer over SSE. Agent / non-TTY callers should pass --json so the full {answer, references, session_id, assistant_message_id} envelope is emitted at completion (no partial chunks). Pass --session to thread follow-ups. Errors: server.session_create_failed when auto-create fails; local.sse_stream_aborted on mid-stream disconnect.")
|
||||
return cmd
|
||||
}
|
||||
|
||||
@@ -229,6 +235,7 @@ func runChat(ctx context.Context, opts *Options, jopts *cmdutil.JSONOptions, svc
|
||||
data := chatData{
|
||||
Answer: answer,
|
||||
References: references,
|
||||
Thinking: acc.Thinking(),
|
||||
SessionID: sid,
|
||||
AssistantMessageID: acc.AssistantMessageID,
|
||||
KBID: opts.KBID,
|
||||
|
||||
Reference in New Issue
Block a user