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.
+
+454. **`claw ls` is caught as a typo with `Did you mean skills` because `"skills".contains("ls") == true`, but the obvious sibling `claw list` falls through to `CliAction::Prompt` and gets shipped to Claude as a user prompt — `suggest_similar_subcommand` at `rust/crates/rusty-claude-cli/src/main.rs:1324-1364` has a 17-entry `KNOWN_SUBCOMMANDS` list, and `list` has Levenshtein distance ≥ 3 from every entry, prefix-match < 4 with every entry, and no substring overlap with any entry, so it returns `None` and the typo-guard arm at `:983-994` is skipped entirely** — dogfooded 2026-05-24 for the 06:30 Clawhip pinpoint nudge at message `1507994083769974825`, reproduced on local `./rust/target/debug/claw` `git_sha 003b739d` (origin/main `f8e1bb72`; the `suggest_similar_subcommand` body is unchanged in `63ce483c..f8e1bb72` which are docs-only ROADMAP additions). Repros in a fully clean isolated environment (`HOME=/tmp/iso4/home` with `{}` settings, fresh `/tmp/iso4/proj` git-init'd workspace, `stdin=/dev/null`, `ANTHROPIC_*` env vars unset): `claw ls </dev/null` exits `1` with `[error-kind: unknown] error: unknown subcommand: ls.\n  Did you mean     skills` (good typed error), while `claw list </dev/null` exits `1` with `[error-kind: missing_credentials] error: missing Anthropic credentials; …` — i.e. the **Anthropic provider** path is invoked because `"list"` was sent as a user prompt instead of being matched by typo suggestion. Adjacent natural CLI-discovery spellings show the same asymmetry: `claw run hello` → prompt dispatch; `claw exec hello` → prompt dispatch; `claw ask what` → prompt dispatch; `claw ls` → typed typo error. The current `KNOWN_SUBCOMMANDS` set (`help`, `version`, `status`, `sandbox`, `doctor`, `state`, `dump-manifests`, `bootstrap-plan`, `agents`, `mcp`, `skills`, `system-prompt`, `acp`, `init`, `export`, `prompt`) is too narrow to catch typos toward CLI discovery shapes that operators reach for from other tools (`ls`, `list`, `run`, `exec`, `ask`, `chat`). The `ls` case only succeeds by coincidence (`"skills".contains("ls")`) — small refactors to the suggestion candidate list could silently regress that and remove the only working typed-error path for the `ls` spelling too. **Why this is distinct from existing items:** #78/#145/#452 each filed a single missing `CliAction` variant (`plugins`, `models`); #453 (PR #3066) filed the slash-command guard being shape-blind to subcommand args. This pinpoint is the **typo-suggestion candidate list** being too narrow to cover natural cross-tool CLI shapes — even spellings that have no corresponding command should at least be caught as typed `unknown subcommand` errors with `did_you_mean` suggestions rather than shipped to the LLM. **Why it matters:** prompt misdelivery is the Clawhip top-of-list category. Operators coming from `gh`, `kubectl`, `ls`, `npm list`, `pip list`, `git log`, `aws ec2 run-instances`, `openai api models.list` will naturally try `claw list`, `claw run …`, `claw exec …`, `claw ask …`. Today, the operator sees `missing_credentials` for an Anthropic prompt when they meant CLI discovery (without creds), or burns provider tokens on a meaningless `"list"`/`"run hello"`/`"ask what"` completion (with creds). The current `ls`→`skills` substring hack is good, but the obvious literal `list` spelling is worse than the typo. **Required fix shape:** (a) extend `KNOWN_SUBCOMMANDS` in `suggest_similar_subcommand` to include natural CLI-discovery shapes that should always be classified as typos: at minimum `list`, `ls`, `models`, `providers`, `profiles`, `aliases`, `logs`, `settings`, `run`, `exec`, `ask`, `chat`, `auth`, `login`, `logout`, `help`, with explicit `did_you_mean` mappings to existing commands or slash commands when applicable (`list`→`skills list` / `mcp list` / `agents list`, `models`→`--model <alias> prompt …` / `/models` once #452 lands, `run`/`exec`→`prompt`, `ask`/`chat`→`prompt`/REPL); (b) keep `looks_like_subcommand_typo` permissive (alphabetic + dash), but for the extended set, attach explicit did-you-mean targets instead of relying on Levenshtein/prefix/substring heuristics that produce coincidental matches like `ls`→`skills`; (c) emit a structured `kind:"unknown_subcommand"` JSON envelope with `input`, `suggestions[]`, and a `prompt_dispatch_blocked:true` flag so automation can distinguish guard rejection from auth failure (mirrors the #453 `argv_misroute_prevented` shape); (d) add regression coverage in `parses_*` test family covering at least `list`, `run`, `exec`, `ask`, `chat`, `models`, `providers`, `profiles`, `aliases`, `logs`, `settings`, asserting every spelling resolves to a typed typo/unknown-subcommand error and **never** to `CliAction::Prompt`. **Acceptance check (one-liner):** `env -u ANTHROPIC_API_KEY -u ANTHROPIC_AUTH_TOKEN claw list` should NOT exit with `missing_credentials`; it should exit with a typed `unknown subcommand: list` error with `Did you mean skills list / mcp list / agents list?`. Source: gaebal-gajae dogfood follow-up for the 2026-05-24 06:30 Clawhip pinpoint nudge at message `1507994083769974825`.
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.

		454. `claw ls` is caught as a typo with `Did you mean skills` because `"skills".contains("ls") == true`, but the obvious sibling `claw list` falls through to `CliAction::Prompt` and gets shipped to Claude as a user prompt — `suggest_similar_subcommand` at `rust/crates/rusty-claude-cli/src/main.rs:1324-1364` has a 17-entry `KNOWN_SUBCOMMANDS` list, and `list` has Levenshtein distance ≥ 3 from every entry, prefix-match < 4 with every entry, and no substring overlap with any entry, so it returns `None` and the typo-guard arm at `:983-994` is skipped entirely — dogfooded 2026-05-24 for the 06:30 Clawhip pinpoint nudge at message `1507994083769974825`, reproduced on local `./rust/target/debug/claw` `git_sha 003b739d` (origin/main `f8e1bb72`; the `suggest_similar_subcommand` body is unchanged in `63ce483c..f8e1bb72` which are docs-only ROADMAP additions). Repros in a fully clean isolated environment (`HOME=/tmp/iso4/home` with `{}` settings, fresh `/tmp/iso4/proj` git-init'd workspace, `stdin=/dev/null`, `ANTHROPIC_` env vars unset): `claw ls </dev/null` exits `1` with `[error-kind: unknown] error: unknown subcommand: ls.\n Did you mean skills` (good typed error), while `claw list </dev/null` exits `1` with `[error-kind: missing_credentials] error: missing Anthropic credentials; …` — i.e. the Anthropic provider* path is invoked because `"list"` was sent as a user prompt instead of being matched by typo suggestion. Adjacent natural CLI-discovery spellings show the same asymmetry: `claw run hello` → prompt dispatch; `claw exec hello` → prompt dispatch; `claw ask what` → prompt dispatch; `claw ls` → typed typo error. The current `KNOWN_SUBCOMMANDS` set (`help`, `version`, `status`, `sandbox`, `doctor`, `state`, `dump-manifests`, `bootstrap-plan`, `agents`, `mcp`, `skills`, `system-prompt`, `acp`, `init`, `export`, `prompt`) is too narrow to catch typos toward CLI discovery shapes that operators reach for from other tools (`ls`, `list`, `run`, `exec`, `ask`, `chat`). The `ls` case only succeeds by coincidence (`"skills".contains("ls")`) — small refactors to the suggestion candidate list could silently regress that and remove the only working typed-error path for the `ls` spelling too. Why this is distinct from existing items: #78/#145/#452 each filed a single missing `CliAction` variant (`plugins`, `models`); #453 (PR #3066) filed the slash-command guard being shape-blind to subcommand args. This pinpoint is the typo-suggestion candidate list being too narrow to cover natural cross-tool CLI shapes — even spellings that have no corresponding command should at least be caught as typed `unknown subcommand` errors with `did_you_mean` suggestions rather than shipped to the LLM. Why it matters: prompt misdelivery is the Clawhip top-of-list category. Operators coming from `gh`, `kubectl`, `ls`, `npm list`, `pip list`, `git log`, `aws ec2 run-instances`, `openai api models.list` will naturally try `claw list`, `claw run …`, `claw exec …`, `claw ask …`. Today, the operator sees `missing_credentials` for an Anthropic prompt when they meant CLI discovery (without creds), or burns provider tokens on a meaningless `"list"`/`"run hello"`/`"ask what"` completion (with creds). The current `ls`→`skills` substring hack is good, but the obvious literal `list` spelling is worse than the typo. Required fix shape: (a) extend `KNOWN_SUBCOMMANDS` in `suggest_similar_subcommand` to include natural CLI-discovery shapes that should always be classified as typos: at minimum `list`, `ls`, `models`, `providers`, `profiles`, `aliases`, `logs`, `settings`, `run`, `exec`, `ask`, `chat`, `auth`, `login`, `logout`, `help`, with explicit `did_you_mean` mappings to existing commands or slash commands when applicable (`list`→`skills list` / `mcp list` / `agents list`, `models`→`--model <alias> prompt …` / `/models` once #452 lands, `run`/`exec`→`prompt`, `ask`/`chat`→`prompt`/REPL); (b) keep `looks_like_subcommand_typo` permissive (alphabetic + dash), but for the extended set, attach explicit did-you-mean targets instead of relying on Levenshtein/prefix/substring heuristics that produce coincidental matches like `ls`→`skills`; (c) emit a structured `kind:"unknown_subcommand"` JSON envelope with `input`, `suggestions[]`, and a `prompt_dispatch_blocked:true` flag so automation can distinguish guard rejection from auth failure (mirrors the #453 `argv_misroute_prevented` shape); (d) add regression coverage in `parses_` test family covering at least `list`, `run`, `exec`, `ask`, `chat`, `models`, `providers`, `profiles`, `aliases`, `logs`, `settings`, asserting every spelling resolves to a typed typo/unknown-subcommand error and never* to `CliAction::Prompt`. Acceptance check (one-liner): `env -u ANTHROPIC_API_KEY -u ANTHROPIC_AUTH_TOKEN claw list` should NOT exit with `missing_credentials`; it should exit with a typed `unknown subcommand: list` error with `Did you mean skills list / mcp list / agents list?`. Source: gaebal-gajae dogfood follow-up for the 2026-05-24 06:30 Clawhip pinpoint nudge at message `1507994083769974825`.