sidekick-agent-hub 0.26.1 → 0.26.3

package/README.md

@@ -8,6 +8,9 @@ Sidekick CLI reads from `~/.config/sidekick/` — the same data files the [VS Co
 
 ## What's New
 
+- **0.26.3: provider failure diagnosis and status evidence** — AI summary failures distinguish authentication, provider sessions, service, connection, timeout, rate-limit, execution-policy, runtime, and context-limit problems with provider-specific guidance, without signing in, switching providers, or replaying. `sidekick status`, the dashboard, and `doctor` show unavailable or partial public status instead of implying normal operation, and `status --json` adds `serviceStatus` with an `availability` discriminator.
+- **0.26.2: reliable dashboard replay** — complete-line checkpoints preserve Unicode and parser context. Live-only runs leave complete-history caches intact; OpenCode replays history when requested.
+- **Project search and Doctor** — project-scoped searches include Codex and database-only OpenCode sessions; `doctor --provider` focuses diagnostics on the selected session provider.
 - **Usage reports from session logs** — `sidekick daily`, `weekly`, `monthly`, and `sessions` compute tokens and cost for every provider straight from session logs (local calendar days, `--breakdown`, `--by-project`, `--csv`, `--json`), so CLI-only users no longer need the extension's history store; `sidekick import` backfills that store when you want `stats` and `today` to see older sessions.
 - **`sidekick blocks`** — five-hour billing blocks with cache-inclusive totals, cost provenance, burn rate, and end-of-block projections, with the official status-line sample beside the local estimate.
 - **Official quota through the status line** — `sidekick statusline` reads the JSON Claude Code pipes on stdin, appends context %, session cost, and cache hit rate, and persists the official limits; `quota`, `quota --all`, and `mcp get_quota_status` share one resolution order and print a `Source` row naming where the numbers came from.
@@ -24,6 +27,10 @@ Sidekick CLI reads from `~/.config/sidekick/` — the same data files the [VS Co
 
 See the [full changelog](https://github.com/cesarandreslopez/sidekick-agent-hub/blob/main/CHANGELOG.md) for everything.
 
+## Replay and Recovery
+
+Selecting a session in the picker loads its history. When starting with `--session <id>`, add `--replay` to include earlier events; without it, the dashboard follows new activity only. Claude Code and Codex can reuse compatible complete-history checkpoints. Live-only runs never replace those checkpoints, and OpenCode replays from the start when history is requested. Doctor honors `--provider` and treats unused integrations as informational.
+
 ## Installation
 
 > **Note:** The npm package is `sidekick-agent-hub`, but the binary it installs is called `sidekick`.
@@ -240,7 +247,9 @@ sidekick extract -i
 sidekick status
 ```
 
-Check API health for both Claude (status.claude.com) and OpenAI (status.openai.com). Shows indicators with color coding (green/yellow/red), affected components, and active incident details. Use `--json` for machine-readable output. In the dashboard, provider-status surfaces are scoped to the monitored provider: Claude for Claude Code sessions, OpenAI for Codex sessions, and hidden for OpenCode.
+Check public service status for both Claude (status.claude.com) and OpenAI (status.openai.com). Shows indicators with color coding (green/yellow/red), component associations, every unresolved incident the feed supplies, and separate check and provider-update timestamps. A failed check displays **Status unavailable** instead of implying normal operation; a feed that omits incidents displays **Incident information unavailable**. Public incidents do not establish why a particular request failed. In the dashboard, provider-status surfaces are scoped to the monitored provider: Claude for Claude Code sessions, OpenAI for Codex sessions, and hidden for OpenCode. Dashboard details and Doctor use the same evidence semantics.
+
+`sidekick status --json` retains `claude`, `openai`, and `peak` for compatibility and adds `serviceStatus: { claude, openai }`. New integrations should use each result's `availability` discriminator; the legacy `indicator: none` fallback cannot distinguish operational status from failed fetching. Doctor similarly adds `serviceStatus` alongside its legacy `providerStatus` field.
 
 When the active provider is `claude-code`, the output also includes a **Claude Peak Hours** block (see below).