ultraworkers · Yeachan-Heo · May 24, 2026
diff --git a/ROADMAP.md b/ROADMAP.md
@@ -6428,3 +6428,5 @@ Original filing (2026-04-18): the session emitted `SessionStart hook (completed)
 
 
 450. **`prompt` emits `kind:"missing_credentials"` JSON on STDERR (not stdout), leaving stdout at 0 bytes — automation pattern `output=$(claw prompt hello --output-format json)` captures nothing on auth-absent failure; `doctor` correctly surfaces `auth.status:"warn"` with `api_key_present:false` but exposes no `prompt_ready:false` field that automation can check before invoking `prompt`** — dogfooded 2026-05-16 by Jobdori on `a35ee9a0` in response to Clawhip pinpoint nudge at `1505208225321062521`. Exact reproduction (isolated env, no creds, fresh git repo, HEAD `a35ee9a0`): `timeout 5 env -i HOME=$ISOLATED_HOME PATH=$PATH CLAW_CONFIG_HOME=$PROBE/.claw-cfg claw prompt hello --output-format json > stdout.txt 2> stderr.txt` → stdout = **0 bytes**, stderr = 195 bytes containing `{"error":"missing Anthropic credentials…","exit_code":1,"hint":null,"kind":"missing_credentials","type":"error"}`, exit code 1. Confirms Gaebal's `1505208553793781792` pinpoint that `prompt` timeout + zero bytes was the prior state — HEAD `a35ee9a0` now correctly exits 1 with `kind:"missing_credentials"` **but the envelope is still routed to stderr** (issue #447 class, same class as prior entries #422, #435). **Contrast with `doctor`:** `claw doctor --output-format json 2>/dev/null` succeeds to stdout with `checks[auth].status:"warn"`, `api_key_present:false`, `auth_token_present:false` — but the auth check has no `prompt_ready:false` field. Automation that gates on `doctor` before invoking `prompt` must re-derive readiness from `api_key_present && auth_token_present` — there is no single canonical boolean. **Three compound problems:** (a) **stdout-empty on `--output-format json` failure**: same class as #447; `prompt`'s error envelope goes to stderr, not stdout. The canonical automation idiom `if ! result=$(claw prompt "q" --output-format json); then echo "$result" | jq .kind; fi` sees `$result=""` on failure — the jq call gets nothing. All `--output-format json` error paths must route JSON to stdout per #447 contract; (b) **`doctor` missing `prompt_ready` field**: `doctor --output-format json` already knows auth is absent (`api_key_present:false`) but surfaces no derived `prompt_ready:bool` or `prompt_blocked_reason:string` field. Automation must infer readiness from `api_key_present || auth_token_present || legacy_*_present` — a 5-field OR across legacy fields that is fragile as auth mechanisms evolve. A single `prompt_ready:false` (with `prompt_blocked_reason:"auth_missing"`) inside the `auth` check would give downstream a stable contract; (c) **`claw prompt` with no auth does no preflight and fires straight at the API**: the preflight check that `doctor` runs (auth discovery) is not reused by `prompt` to emit a fast typed error before attempting the network call. Both Gaebal's pinpoint (prompt hanging silently on older HEAD) and the current behavior (prompt hitting auth gate after a brief API attempt) stem from the same root: prompt does not short-circuit at the point where `doctor` already knows auth is absent. If `doctor` can emit `kind:"doctor"` with `auth.status:"warn"` in ~20ms without a network call, `prompt` should emit `kind:"missing_credentials"` in the same window and output it to stdout. **Required fix shape:** (a) `prompt --output-format json` must write the `kind:"missing_credentials"` JSON envelope to **stdout**, not stderr — same fix as #447 for all error envelopes; (b) add `prompt_ready:bool` and `prompt_blocked_reason:string|null` to the `auth` check in `doctor --output-format json`; derive it as `api_key_present || auth_token_present || legacy_saved_oauth_present`; (c) `prompt` must run the credential preflight check (same codepath as doctor's auth check) before attempting any API call and emit `{"kind":"missing_credentials","prompt_blocked_reason":"auth_missing"}` on **stdout** with exit 1 if the check fails; (d) `--output-format json` stdout routing fix must cover: `prompt`, `session list` (cross-ref #449), `skills uninstall` (cross-ref #431), `resume` (cross-ref #435), `acp serve` (cross-ref #443) — the full `kind:"missing_credentials"` class; (e) regression test: `claw prompt hello --output-format json` with no creds writes JSON to stdout (0 bytes stderr), exits 1, `kind:"missing_credentials"`, in under 200ms (no network attempt). **Why this matters:** `prompt` is the primary consumer entry point. Auth-absent failure routing to stderr breaks every automation wrapper that captures `$(claw prompt ... --output-format json)`. The `doctor` preflight metadata gap means auth-readiness checks require parsing 5 legacy fields instead of reading one boolean. Cross-references #447 (all JSON error envelopes on stderr), #449 (session list hits auth gate), #431 (skills uninstall hits auth gate), #357 (auth gate on local ops cluster), #422 (exit-code parity). Source: Jobdori live dogfood, `a35ee9a0`, 2026-05-16.
+
+453. **`bare_slash_command_guidance` (the "is a slash command" guard) only fires for `command_name` matched as a single bare token at `rust/crates/rusty-claude-cli/src/main.rs:1100-1149` — as soon as ANY positional argument follows a slash-command-named first token, the parser falls through to `CliAction::Prompt` and ships the entire argv string to Claude as a user prompt. The guard catches `claw cost`, `claw tokens`, `claw model`, `claw permissions`, `claw context`, `claw providers`, `claw history`, `claw release-notes`, `claw review`, `claw compact`, `claw cache` — but misses `claw cost list`, `claw tokens list`, `claw model openai/gpt-4`, `claw model list`, `claw permissions show`, `claw context show`, `claw providers list`, `claw history list`, `claw cache list`. Same pattern bites unrouted command shapes: `claw aliases`, `claw aliases list`, `claw profiles list`, `claw logs`, `claw settings` all fall through, even though they are obvious CLI discovery spellings** — dogfooded 2026-05-24 for the 06:00 Clawhip pinpoint nudge at message `1507986539538419863`, reproduced on local `./rust/target/debug/claw` `git_sha 003b739d` (origin/main `f8e1bb72`; the guard logic in `bare_slash_command_guidance` was last touched by the `#146` `config`/`diff` carve-out and is unchanged in `63ce483c..f8e1bb72` which are docs-only ROADMAP additions). Repros in a fully clean isolated environment (`HOME=/tmp/iso3/home` with `{}` settings, fresh `/tmp/iso3/proj` git-init'd workspace, `stdin=/dev/null`, `ANTHROPIC_*` env vars unset): bare `claw cost </dev/null` exits `1` with `[error-kind: unknown] error: `claw cost` is a slash command. …`; adding a single argument `claw cost list </dev/null` exits `1` with `[error-kind: missing_credentials] error: missing Anthropic credentials; …` — i.e. the **Anthropic provider** path is invoked because `"cost list"` was sent as a user prompt instead of being caught by the guard. Same observation for `claw model openai/gpt-4` (operator-natural typo for setting a model), `claw permissions show`, `claw context show`, and all the unrouted plural-noun spellings above. This is materially broader than #78/#145/#452: those filed `plugins` and `models` as **single missing `CliAction` variants**; the pinpoint here is the **guard itself** being shape-blind to subcommand args, so even commands whose bare form *is* correctly handled regress to prompt misdelivery the moment an operator types one more word — including `--model <alias>` confusion (`claw model openai/gpt-4`) which is an extremely natural typo given the existing `claw --model openai/gpt-4 prompt …` flag shape. **Why this is distinct from existing items:** #78/#145 covered `claw plugins` only; #452 covered `claw models*` only. #357 / #322 cover JSON envelope/stderr-prefix issues, not argv classification. The guard surface lives in `parse_subcommand` and `bare_slash_command_guidance` (`rust/crates/rusty-claude-cli/src/main.rs:1100-1149`) — no existing ROADMAP entry tracks the shape-blindness regression class across the guard. **Why it matters:** prompt misdelivery is the Clawhip top-of-list category. With credentials set, every operator/claw typo in this cluster burns provider tokens on a meaningless completion (e.g., sending `"cost list"` to Claude). Without credentials, the wrong-shaped `missing_credentials` envelope is even more confusing — operators see `missing Anthropic credentials` when they meant local cost inspection, and assume an auth bug rather than a CLI dispatch bug. The fix is also extremely localized: it lives in a single guard function. **Required fix shape:** (a) widen `bare_slash_command_guidance` to also fire from the subcommand-args parse arm: when the first positional token matches a known slash-command name and any additional args follow, emit a typed guard error (`"`claw cost list` is a slash command — `claw cost` does not accept extra arguments; use `claw --resume SESSION.jsonl /cost` instead"`) before falling through to `CliAction::Prompt`; (b) extend the guard's known-slash-command set to include the natural CLI discovery spellings that currently fall through entirely (`models`, `model`, `aliases`, `profiles`, `providers list`, `logs`, `settings`), even when there is no corresponding slash command — emit a typed `"unknown CLI subcommand"` error with `did_you_mean` suggestions (`--model <alias> prompt …`, `/models`, `/providers`, `claw config plugins list`) rather than dispatching to LLM prompt; (c) add a structured `kind:"argv_misroute_prevented"` JSON envelope so automation can distinguish guard rejection from auth failure; (d) add regression coverage in `parses_*` test family covering at least one extra-arg case per existing slash-command guard (`cost list`, `tokens list`, `model list`, `model openai/gpt-4`, `permissions show`, `context show`, `cache list`) plus the unrouted-noun cluster (`models list`, `aliases`, `profiles list`, `logs`), asserting every spelling resolves to a typed guard error and **never** to `CliAction::Prompt`. **Acceptance check (one-liner):** `env -u ANTHROPIC_API_KEY -u ANTHROPIC_AUTH_TOKEN claw model openai/gpt-4` should NOT exit with `missing_credentials`; it should exit with a typed CLI dispatch error mentioning `--model openai/gpt-4 prompt …` or `/model`. Source: gaebal-gajae dogfood follow-up for the 2026-05-24 06:00 Clawhip pinpoint nudge at message `1507986539538419863`.
Original file line number	Diff line number	Diff line change
Expand Up		@@ -6428,3 +6428,5 @@ Original filing (2026-04-18): the session emitted `SessionStart hook (completed)


		450. `prompt` emits `kind:"missing_credentials"` JSON on STDERR (not stdout), leaving stdout at 0 bytes — automation pattern `output=$(claw prompt hello --output-format json)` captures nothing on auth-absent failure; `doctor` correctly surfaces `auth.status:"warn"` with `api_key_present:false` but exposes no `prompt_ready:false` field that automation can check before invoking `prompt` — dogfooded 2026-05-16 by Jobdori on `a35ee9a0` in response to Clawhip pinpoint nudge at `1505208225321062521`. Exact reproduction (isolated env, no creds, fresh git repo, HEAD `a35ee9a0`): `timeout 5 env -i HOME=$ISOLATED_HOME PATH=$PATH CLAW_CONFIG_HOME=$PROBE/.claw-cfg claw prompt hello --output-format json > stdout.txt 2> stderr.txt` → stdout = 0 bytes, stderr = 195 bytes containing `{"error":"missing Anthropic credentials…","exit_code":1,"hint":null,"kind":"missing_credentials","type":"error"}`, exit code 1. Confirms Gaebal's `1505208553793781792` pinpoint that `prompt` timeout + zero bytes was the prior state — HEAD `a35ee9a0` now correctly exits 1 with `kind:"missing_credentials"` but the envelope is still routed to stderr (issue #447 class, same class as prior entries #422, #435). Contrast with `doctor`: `claw doctor --output-format json 2>/dev/null` succeeds to stdout with `checks[auth].status:"warn"`, `api_key_present:false`, `auth_token_present:false` — but the auth check has no `prompt_ready:false` field. Automation that gates on `doctor` before invoking `prompt` must re-derive readiness from `api_key_present && auth_token_present` — there is no single canonical boolean. Three compound problems: (a) stdout-empty on `--output-format json` failure: same class as #447; `prompt`'s error envelope goes to stderr, not stdout. The canonical automation idiom `if ! result=$(claw prompt "q" --output-format json); then echo "$result" \| jq .kind; fi` sees `$result=""` on failure — the jq call gets nothing. All `--output-format json` error paths must route JSON to stdout per #447 contract; (b) `doctor` missing `prompt_ready` field: `doctor --output-format json` already knows auth is absent (`api_key_present:false`) but surfaces no derived `prompt_ready:bool` or `prompt_blocked_reason:string` field. Automation must infer readiness from `api_key_present \|\| auth_token_present \|\| legacy__present` — a 5-field OR across legacy fields that is fragile as auth mechanisms evolve. A single `prompt_ready:false` (with `prompt_blocked_reason:"auth_missing"`) inside the `auth` check would give downstream a stable contract; (c) `claw prompt` with no auth does no preflight and fires straight at the API: the preflight check that `doctor` runs (auth discovery) is not reused by `prompt` to emit a fast typed error before attempting the network call. Both Gaebal's pinpoint (prompt hanging silently on older HEAD) and the current behavior (prompt hitting auth gate after a brief API attempt) stem from the same root: prompt does not short-circuit at the point where `doctor` already knows auth is absent. If `doctor` can emit `kind:"doctor"` with `auth.status:"warn"` in ~20ms without a network call, `prompt` should emit `kind:"missing_credentials"` in the same window and output it to stdout. Required fix shape:* (a) `prompt --output-format json` must write the `kind:"missing_credentials"` JSON envelope to stdout, not stderr — same fix as #447 for all error envelopes; (b) add `prompt_ready:bool` and `prompt_blocked_reason:string\|null` to the `auth` check in `doctor --output-format json`; derive it as `api_key_present \|\| auth_token_present \|\| legacy_saved_oauth_present`; (c) `prompt` must run the credential preflight check (same codepath as doctor's auth check) before attempting any API call and emit `{"kind":"missing_credentials","prompt_blocked_reason":"auth_missing"}` on stdout with exit 1 if the check fails; (d) `--output-format json` stdout routing fix must cover: `prompt`, `session list` (cross-ref #449), `skills uninstall` (cross-ref #431), `resume` (cross-ref #435), `acp serve` (cross-ref #443) — the full `kind:"missing_credentials"` class; (e) regression test: `claw prompt hello --output-format json` with no creds writes JSON to stdout (0 bytes stderr), exits 1, `kind:"missing_credentials"`, in under 200ms (no network attempt). Why this matters: `prompt` is the primary consumer entry point. Auth-absent failure routing to stderr breaks every automation wrapper that captures `$(claw prompt ... --output-format json)`. The `doctor` preflight metadata gap means auth-readiness checks require parsing 5 legacy fields instead of reading one boolean. Cross-references #447 (all JSON error envelopes on stderr), #449 (session list hits auth gate), #431 (skills uninstall hits auth gate), #357 (auth gate on local ops cluster), #422 (exit-code parity). Source: Jobdori live dogfood, `a35ee9a0`, 2026-05-16.

		453. `bare_slash_command_guidance` (the "is a slash command" guard) only fires for `command_name` matched as a single bare token at `rust/crates/rusty-claude-cli/src/main.rs:1100-1149` — as soon as ANY positional argument follows a slash-command-named first token, the parser falls through to `CliAction::Prompt` and ships the entire argv string to Claude as a user prompt. The guard catches `claw cost`, `claw tokens`, `claw model`, `claw permissions`, `claw context`, `claw providers`, `claw history`, `claw release-notes`, `claw review`, `claw compact`, `claw cache` — but misses `claw cost list`, `claw tokens list`, `claw model openai/gpt-4`, `claw model list`, `claw permissions show`, `claw context show`, `claw providers list`, `claw history list`, `claw cache list`. Same pattern bites unrouted command shapes: `claw aliases`, `claw aliases list`, `claw profiles list`, `claw logs`, `claw settings` all fall through, even though they are obvious CLI discovery spellings — dogfooded 2026-05-24 for the 06:00 Clawhip pinpoint nudge at message `1507986539538419863`, reproduced on local `./rust/target/debug/claw` `git_sha 003b739d` (origin/main `f8e1bb72`; the guard logic in `bare_slash_command_guidance` was last touched by the `#146` `config`/`diff` carve-out and is unchanged in `63ce483c..f8e1bb72` which are docs-only ROADMAP additions). Repros in a fully clean isolated environment (`HOME=/tmp/iso3/home` with `{}` settings, fresh `/tmp/iso3/proj` git-init'd workspace, `stdin=/dev/null`, `ANTHROPIC_` env vars unset): bare `claw cost </dev/null` exits `1` with `[error-kind: unknown] error: `claw cost` is a slash command. …`; adding a single argument `claw cost list </dev/null` exits `1` with `[error-kind: missing_credentials] error: missing Anthropic credentials; …` — i.e. the Anthropic provider* path is invoked because `"cost list"` was sent as a user prompt instead of being caught by the guard. Same observation for `claw model openai/gpt-4` (operator-natural typo for setting a model), `claw permissions show`, `claw context show`, and all the unrouted plural-noun spellings above. This is materially broader than #78/#145/#452: those filed `plugins` and `models` as single missing `CliAction` variants; the pinpoint here is the guard itself being shape-blind to subcommand args, so even commands whose bare form is correctly handled regress to prompt misdelivery the moment an operator types one more word — including `--model <alias>` confusion (`claw model openai/gpt-4`) which is an extremely natural typo given the existing `claw --model openai/gpt-4 prompt …` flag shape. Why this is distinct from existing items: #78/#145 covered `claw plugins` only; #452 covered `claw models` only. #357 / #322 cover JSON envelope/stderr-prefix issues, not argv classification. The guard surface lives in `parse_subcommand` and `bare_slash_command_guidance` (`rust/crates/rusty-claude-cli/src/main.rs:1100-1149`) — no existing ROADMAP entry tracks the shape-blindness regression class across the guard. Why it matters:* prompt misdelivery is the Clawhip top-of-list category. With credentials set, every operator/claw typo in this cluster burns provider tokens on a meaningless completion (e.g., sending `"cost list"` to Claude). Without credentials, the wrong-shaped `missing_credentials` envelope is even more confusing — operators see `missing Anthropic credentials` when they meant local cost inspection, and assume an auth bug rather than a CLI dispatch bug. The fix is also extremely localized: it lives in a single guard function. Required fix shape: (a) widen `bare_slash_command_guidance` to also fire from the subcommand-args parse arm: when the first positional token matches a known slash-command name and any additional args follow, emit a typed guard error (`"`claw cost list` is a slash command — `claw cost` does not accept extra arguments; use `claw --resume SESSION.jsonl /cost` instead"`) before falling through to `CliAction::Prompt`; (b) extend the guard's known-slash-command set to include the natural CLI discovery spellings that currently fall through entirely (`models`, `model`, `aliases`, `profiles`, `providers list`, `logs`, `settings`), even when there is no corresponding slash command — emit a typed `"unknown CLI subcommand"` error with `did_you_mean` suggestions (`--model <alias> prompt …`, `/models`, `/providers`, `claw config plugins list`) rather than dispatching to LLM prompt; (c) add a structured `kind:"argv_misroute_prevented"` JSON envelope so automation can distinguish guard rejection from auth failure; (d) add regression coverage in `parses_` test family covering at least one extra-arg case per existing slash-command guard (`cost list`, `tokens list`, `model list`, `model openai/gpt-4`, `permissions show`, `context show`, `cache list`) plus the unrouted-noun cluster (`models list`, `aliases`, `profiles list`, `logs`), asserting every spelling resolves to a typed guard error and never* to `CliAction::Prompt`. Acceptance check (one-liner): `env -u ANTHROPIC_API_KEY -u ANTHROPIC_AUTH_TOKEN claw model openai/gpt-4` should NOT exit with `missing_credentials`; it should exit with a typed CLI dispatch error mentioning `--model openai/gpt-4 prompt …` or `/model`. Source: gaebal-gajae dogfood follow-up for the 2026-05-24 06:00 Clawhip pinpoint nudge at message `1507986539538419863`.