docs(roadmap): add #360 — tokens/stats/cost identical output; no context-window or cost_usd

ROADMAP.md

@@ -6289,4 +6289,4 @@ Original filing (2026-04-18): the session emitted `SessionStart hook (completed)
 347. **Top-level `mcp show <missing-server> --output-format json` reports a missing server as `status:"ok"` instead of a typed not-found/error status** — dogfooded 2026-04-29 for the 20:30 nudge on current `origin/main` / rebuilt `./rust/target/debug/claw` with embedded `git_sha` `ee41b266`. After rebuilding and verifying the binary provenance, running `./rust/target/debug/claw mcp show does-not-exist --output-format json` returned stdout JSON with `{"action":"show","config_load_error":null,"found":false,"kind":"mcp","message":"server `does-not-exist` is not configured","server_name":"does-not-exist","status":"ok"}` and no stderr. `found:false` is useful, but pairing it with `status:"ok"` makes the command-level outcome ambiguous: a missing requested server is not an OK inspection result for automation that needs to distinguish successful detail retrieval from a not-found lookup. This is distinct from #327's MCP source-list mismatch and the invalid #2874/#2879/#2880 hang/nondeterminism hypotheses that were closed after bounded repros. **Required fix shape:** (a) return a typed not-found status such as `status:"not_found"` or `kind:"error"` plus `code:"mcp_server_not_found"` while preserving `server_name` and optional `available_servers[]`; (b) document whether `found:false` objects are considered success or error and keep that convention consistent across text and JSON modes; (c) ensure process exit semantics match the JSON status contract or expose a separate `exit_ok`/`lookup_status` field; (d) add regression coverage proving missing-server lookup is distinguishable from successful server detail retrieval without parsing the human `message`. **Why this matters:** MCP inspection is a control-plane diagnostic. If a missing server returns `status:"ok"`, claws can silently treat a failed lookup as healthy MCP state unless they special-case `found:false`, which defeats the purpose of a clear machine-readable status field. Source: gaebal-gajae dogfood follow-up for the 20:30 nudge on rebuilt `./rust/target/debug/claw` `ee41b266`.
 348. **Top-level `plugins list --output-format json` returns plugin inventory only as a prose `message` string instead of structured `plugins[]` entries** — dogfooded 2026-04-29 for the 21:00 nudge on current `origin/main` / rebuilt `./rust/target/debug/claw` with embedded `git_sha` `cca6f682`. Running `./rust/target/debug/claw plugins list --output-format json` repeatedly returned valid stdout JSON with `{"action":"list","kind":"plugin","message":"Plugins\n  example-bundled      v0.1.0      disabled\n  sample-hooks         v0.1.0      disabled","reload_runtime":false,"target":null}` and no stderr. The actual plugin names, versions, and enabled/disabled states are present only inside the human-formatted `message` table; there is no `plugins[]` array, no per-plugin `name`, `version`, `enabled`, `source`, `load_error`, or lifecycle/action metadata. This is distinct from #325's broad help JSON opacity and the config/MCP/agent items: the affected surface is plugin lifecycle inventory, where automation needs a structured list before enabling, disabling, updating, or uninstalling plugins. **Required fix shape:** (a) add `plugins[]` with stable per-plugin fields such as `name`, `version`, `enabled`, `source`, `configured`, `load_status`, and optional `error`; (b) keep `message` only as a human summary, not the sole inventory payload; (c) expose counts and truncation metadata if the list can be large; (d) add regression coverage proving `plugins list --output-format json` can be parsed without scraping the prose message and that disabled/enabled state survives as booleans/enums. **Why this matters:** plugin lifecycle management is a control-plane path. If the JSON inventory is just a text table, claws must scrape spacing-sensitive prose before deciding whether a plugin is installed, disabled, broken, or safe to mutate. Source: gaebal-gajae dogfood follow-up for the 21:00 nudge on rebuilt `./rust/target/debug/claw` `cca6f682`.
 
-361. **`session_not_found` error includes recovery information only as a prose `hint` string; the sessions directory path, workspace partition key, and session list command are not structured fields** — dogfooded 2026-04-30 by Jobdori on `a2a38df9`. Running `./rust/target/debug/claw --output-format json --resume /nonexistent/path.jsonl /status` returns: `{"error":"failed to restore session: session not found: /nonexistent/path.jsonl","hint":"Hint: managed sessions live in .claw/sessions/4dbe3d911e02dd59/ (workspace-specific partition).\nTry 'latest' for the most recent session or '/session list' in the REPL.","kind":"session_not_found","type":"error"}`. The `hint` field contains the sessions directory path, the workspace partition key, and the recovery action as a human-formatted prose string with embedded newlines. An orchestrator recovering from a session-not-found error must: (a) regex-parse the hint to extract the sessions directory (`/.claw/sessions/<partition>/`), (b) infer the workspace partition key from the prose, and (c) handle the `/session list` suggestion even though that command requires valid Anthropic credentials (see #357). None of this is available as a structured field. This is distinct from #357's credential requirement for `session list` and #356's null title: the pinpoint here is that the error recovery path for a missing session exposes machine-critical data only in prose format. **Required fix shape:** (a) add `sessions_dir` field with the absolute path to the sessions directory; (b) add `workspace_partition` field with the session partition key; (c) add `resolution` object with `kind:"use_latest"` and the canonical `--resume latest` form; (d) keep `hint` for human display but ensure all actionable recovery data has a structured parallel; (e) add regression coverage confirming `session_not_found` includes non-null `sessions_dir` and `workspace_partition`. Source: Jobdori live dogfood, mengmotaHost, `a2a38df9`, 2026-04-30.
+360. **`/tokens`, `/stats`, and `/cost --output-format json` return identical token-count blobs differing only in `kind`; none exposes `context_window`, `context_remaining`, or cost in USD** — dogfooded 2026-04-30 by Jobdori on `c6c01bea`. Running all three commands under `--output-format json --resume latest` returns structurally identical responses with the same `input_tokens`, `output_tokens`, `cache_read_tokens`, `cache_write_tokens` values — only the `kind` field differs (`"tokens"`, `"stats"`, `"cost"` respectively). None includes `context_window` (the model's max token limit), `context_remaining` (available headroom), `cost_usd`, `currency`, or `model_id`. An orchestrator tracking context utilisation cannot determine how close the session is to the limit; an orchestrator tracking spend has no dollar amount. The three commands appear to be the same handler with a different `kind` label, providing no additional surface. **Required fix shape:** (a) `/tokens` — add `context_window`, `context_remaining`, `context_utilisation_pct`, and `model_id`; (b) `/cost` — add `cost_usd`, `currency`, and per-type cost breakdown; (c) `/stats` — combine token counts + cost + context window into a session summary; (d) if the commands are intentionally separated, differentiate their payloads so each name delivers its named information rather than returning the same blob; (e) add regression coverage confirming `/tokens` includes `context_window > 0` and `/cost` includes `cost_usd >= 0` for a session with at least one completed turn. Source: Jobdori live dogfood, mengmotaHost, `c6c01bea`, 2026-04-30.