xai-grok-build CLI
2/8 principles met
xAI's Grok Build TUI — agentic coding CLI for Grok models
$ anc audit grok --json
Eight principles, scored
Behavioral and source checks. Every non-pass row carries a copy-paste remediation prompt.
- P1Non-Interactive by Defaultpass
All 6 checks pass.
- P2Structured, Parseable Outputfail
no --json or --jsonl short alias found. Agents and pipelines benefit from short forms alongside the canonical `--output` enum. · no `--raw` flag advertised. MAY-tier — useful for pipelines that want to strip formatting before piping to other tools.
Remediation · P2Make xai-grok-build satisfy P2 (Structured, Parseable Output) from the agent-native CLI standard. Address: - --json / --jsonl short aliases for --output (no --json or --jsonl short alias found. Agents and pipelines benefit from short forms alongside the canonical `--output` enum.) - `--raw` flag for pipe-safe unformatted output (no `--raw` flag advertised. MAY-tier — useful for pipelines that want to strip formatting before piping to other tools.) - Errors emit JSON envelope with discriminant + type id + message roles under `--output json` (bad invocation under `--output-format json` produced no parseable JSON on stderr or stdout. JSON mode must emit a JSON error envelope, not plain text.) Requirements: https://anc.dev/p2
- P3Progressive Help Discoveryfail
no `examples` subcommand or `--examples` flag found. MAY-tier — a curated usage block keeps agents from hunting through long help text. · subcommands missing example invocations in their `--help`: agent, clone, cursor-worker, doctor, du, export, inspect, leader, login, mcp, memory, plugin, sessions, setup, trace, update, usage, version, worktree. Examples teach agents the call shape faster than option tables; use clap's `after_help` or a dedicated `Examples:` block. exempt: dashboard, logout, models (take no arguments).
Remediation · P3Make xai-grok-build satisfy P3 (Progressive Help Discovery) from the agent-native CLI standard. Address: - `examples` subcommand or `--examples` flag for curated usage patterns (no `examples` subcommand or `--examples` flag found. MAY-tier — a curated usage block keeps agents from hunting through long help text.) - Each subcommand that takes arguments ships an invocation example in its `--help` (subcommands missing example invocations in their `--help`: agent, clone, cursor-worker, doctor, du, export, inspect, leader, login, mcp, memory, plugin, sessions, setup, trace, update, usage, version, worktree. Examples teach agents the call shape faster than option tables; use clap's `after_help` or a dedicated `Examples:` block. exempt: dashboard, logout, models (take no arguments).) - Help text pairs human and `--output json` example invocations (no paired text + `--output json` example found within 5 lines in top-level or any subcommand `--help`. Pairing keeps agents from reverse-engineering the JSON invocation from the text one.) Requirements: https://anc.dev/p3
- P4Fail-Fast, Actionable Errorswarn
errors under `--output-format json` are not JSON-formatted. Consumers parsing stdout-as-JSON cannot recover the failure without a separate text-parsing path; switch the error writer to honor the active output mode.
Remediation · P4Make xai-grok-build satisfy P4 (Fail-Fast, Actionable Errors) from the agent-native CLI standard. Address: - `--output json` produces JSON-formatted errors (errors under `--output-format json` are not JSON-formatted. Consumers parsing stdout-as-JSON cannot recover the failure without a separate text-parsing path; switch the error writer to honor the active output mode.) Requirements: https://anc.dev/p4
- P5Safe Retries & Mutation Boundarieswarn
write-pattern subcommand(s) present (update) but no read-pattern surface detected. If the CLI is write-only by design the MUST is satisfied vacuously; otherwise expose the read surface with agent-recognizable verbs (list/get/show/query/find/search).
Remediation · P5Make xai-grok-build satisfy P5 (Safe Retries & Mutation Boundaries) from the agent-native CLI standard. Address: - Read and write surfaces are both visible in subcommand list (write-pattern subcommand(s) present (update) but no read-pattern surface detected. If the CLI is write-only by design the MUST is satisfied vacuously; otherwise expose the read surface with agent-recognizable verbs (list/get/show/query/find/search).) Requirements: https://anc.dev/p5
- P6Composable, Predictable Command Structurewarn
8/25 subcommand(s) follow standard verb names. Non-standard: agent, clone, cursor-worker, dashboard, du, export, leader, mcp, memory, models, plugin, sessions, setup, trace, usage, worktree, wrap. MAY-tier — community-standard verbs (get/list/create/update/delete) help agents predict subcommand behavior across CLIs. · no `--color` flag advertised. MAY-tier — `auto|always|never` lets agents and pipelines override the TTY-based default.
Remediation · P6Make xai-grok-build satisfy P6 (Composable, Predictable Command Structure) from the agent-native CLI standard. Address: - Subcommand verbs follow community-standard names (8/25 subcommand(s) follow standard verb names. Non-standard: agent, clone, cursor-worker, dashboard, du, export, leader, mcp, memory, models, plugin, sessions, setup, trace, usage, worktree, wrap. MAY-tier — community-standard verbs (get/list/create/update/delete) help agents predict subcommand behavior across CLIs.) - `--color` flag for explicit color control (no `--color` flag advertised. MAY-tier — `auto|always|never` lets agents and pipelines override the TTY-based default.) - Subcommand naming follows a consistent verb/noun convention (subcommand naming is inconsistent: 5 non-verb subcommand(s) (agent, leader, mcp, plugin, worktree) mix verb and non-verb children at the second level, so an agent cannot predict where the action lives. SHOULD-tier: pick a consistent shape (all verb-first, all noun-verb hierarchy, or any combination where each non-verb group's children are uniformly verbs). The verb list is a heuristic; inspect `--help` to confirm.) Requirements: https://anc.dev/p6
- P7Bounded, High-Signal Responseswarn
no option definition in --help declares --quiet or -q; a flag named only in a usage line does not count. · no `--verbose` / `-v` flag advertised. SHOULD-tier — agents debugging failures need a way to escalate diagnostic detail. `-v` is declared, but its definition (`-v, --version Print version`) does not mean verbose.
Remediation · P7Make xai-grok-build satisfy P7 (Bounded, High-Signal Responses) from the agent-native CLI standard. Address: - Quiet mode available (no option definition in --help declares --quiet or -q; a flag named only in a usage line does not count.) - `--verbose` flag for diagnostic escalation (no `--verbose` / `-v` flag advertised. SHOULD-tier — agents debugging failures need a way to escalate diagnostic detail. `-v` is declared, but its definition (`-v, --version Print version`) does not mean verbose.) Requirements: https://anc.dev/p7
- P8Discoverable Through Agent Skill Bundlespass
All 3 checks pass.
All Audits
P1: Non-Interactive by Default
| PASS | Non-interactive by default | |
| SKIP | Non-interactive gate flag advertised in --help | target satisfies P1 via alternative gate (help-on-bare or stdin-primary) |
| PASS | Flags advertise env-var bindings in --help | |
| PASS | Secret-bearing flags expose stdin or *-file companion | |
| PASS | `--help` advertises default values for flags | |
| PASS | Rich-TUI affordance for TTY contexts |
P2: Structured, Parseable Output
| SKIP | Structured output support | `--output-format` is declared in --help, but no safe probe printed JSON (--help and --version override output flags in most CLIs) |
| SKIP | Structured-output CLI exposes its schema at runtime | antecedent `p2-json-output` could not be measured: `--output-format` is declared in --help, but no safe probe printed JSON (--help and --version override output flags in most CLIs) |
| WARN | --json / --jsonl short aliases for --output | no --json or --jsonl short alias found. Agents and pipelines benefit from short forms alongside the canonical `--output` enum. |
| WARN | `--raw` flag for pipe-safe unformatted output | no `--raw` flag advertised. MAY-tier — useful for pipelines that want to strip formatting before piping to other tools. |
| SKIP | `--output` advertises additional formats beyond text/json | no `--output` or `--format` flag advertised; vacuous skip for MAY-tier extra formats. |
| PASS | Bad invocation exits with structured usage-error code (2) | |
| FAIL | Errors emit JSON envelope with discriminant + type id + message roles under `--output json` | bad invocation under `--output-format json` produced no parseable JSON on stderr or stdout. JSON mode must emit a JSON error envelope, not plain text. |
| SKIP | JSON success and error envelopes share their non-payload key set | success-mode probe (`--help --output-format json`) produced no parseable JSON; cannot compare envelope shapes. |
P3: Progressive Help Discovery
| PASS | Help flag produces useful output | |
| PASS | Version flag works (`--version` plus short alias) | |
| PASS | Version flag works (`--version` plus short alias) | |
| WARN | `examples` subcommand or `--examples` flag for curated usage patterns | no `examples` subcommand or `--examples` flag found. MAY-tier — a curated usage block keeps agents from hunting through long help text. |
| PASS | Short `-h` summary differs from `--help` long form | |
| FAIL | Each subcommand that takes arguments ships an invocation example in its `--help` | subcommands missing example invocations in their `--help`: agent, clone, cursor-worker, doctor, du, export, inspect, leader, login, mcp, memory, plugin, sessions, setup, trace, update, usage, version, worktree. Examples teach agents the call shape faster than option tables; use clap's `after_help` or a dedicated `Examples:` block. exempt: dashboard, logout, models (take no arguments). |
| WARN | Help text pairs human and `--output json` example invocations | no paired text + `--output json` example found within 5 lines in top-level or any subcommand `--help`. Pairing keeps agents from reverse-engineering the JSON invocation from the text one. |
| PASS | Command-list entries name the command without repeating the binary name |
P4: Fail-Fast, Actionable Errors
| PASS | Rejects invalid arguments | |
| PASS | Error messages include a hint or remediation phrase | |
| WARN | `--output json` produces JSON-formatted errors | errors under `--output-format json` are not JSON-formatted. Consumers parsing stdout-as-JSON cannot recover the failure without a separate text-parsing path; switch the error writer to honor the active output mode. |
P5: Safe Retries & Mutation Boundaries
| SKIP | Destructive subcommands require `--force` or `--yes` | no destructive subcommands detected; MUST applies conditionally to CLIs with destructive operations. |
| WARN | Read and write surfaces are both visible in subcommand list | write-pattern subcommand(s) present (update) but no read-pattern surface detected. If the CLI is write-only by design the MUST is satisfied vacuously; otherwise expose the read surface with agent-recognizable verbs (list/get/show/query/find/search). |
P6: Composable, Predictable Command Structure
| PASS | Handles SIGPIPE gracefully | |
| SKIP | Pager-using CLI ships --no-pager escape hatch | no pager signal ($PAGER, --pager, pager, or less/more named as a program) in --help |
| PASS | Respects NO_COLOR | |
| WARN | Subcommand verbs follow community-standard names | 8/25 subcommand(s) follow standard verb names. Non-standard: agent, clone, cursor-worker, dashboard, du, export, leader, mcp, memory, models, plugin, sessions, setup, trace, usage, worktree, wrap. MAY-tier — community-standard verbs (get/list/create/update/delete) help agents predict subcommand behavior across CLIs. |
| WARN | `--color` flag for explicit color control | no `--color` flag advertised. MAY-tier — `auto|always|never` lets agents and pipelines override the TTY-based default. |
| SKIP | Input-accepting commands read from stdin when no file is given | no input-accepting subcommand detected (process/parse/convert/transform/analyze/validate/format/lint/audit); vacuous skip for the conditional SHOULD. |
| WARN | Subcommand naming follows a consistent verb/noun convention | subcommand naming is inconsistent: 5 non-verb subcommand(s) (agent, leader, mcp, plugin, worktree) mix verb and non-verb children at the second level, so an agent cannot predict where the action lives. SHOULD-tier: pick a consistent shape (all verb-first, all noun-verb hierarchy, or any combination where each non-verb group's children are uniformly verbs). The verb list is a heuristic; inspect `--help` to confirm. |
| PASS | Operations are subcommands, not verb-shaped flags |
P7: Bounded, High-Signal Responses
| WARN | Quiet mode available | no option definition in --help declares --quiet or -q; a flag named only in a usage line does not count. |
| WARN | `--verbose` flag for diagnostic escalation | no `--verbose` / `-v` flag advertised. SHOULD-tier — agents debugging failures need a way to escalate diagnostic detail. `-v` is declared, but its definition (`-v, --version Print version`) does not mean verbose. |
| SKIP | `--limit` / `--max-results` flag for list operations | no list-style subcommand detected (list/ls/search/query/find, or a show/get whose summary describes several results); vacuous skip for the list-only SHOULD. |
| SKIP | Cursor-based pagination flags for list traversal | no list-style subcommand detected (list/ls/search/query/find, or a show/get whose summary describes several results); vacuous skip for the list-only MAY. |
| SKIP | `--timeout` flag for long-running operations | no long-running subcommand detected (serve/daemon/watch/tail/monitor/follow/run/start/stream); vacuous skip for the conditional SHOULD. |
| PASS | Help text advertises TTY-aware verbosity behavior |
P8: Discoverable Through Agent Skill Bundles
| N/A | Skill bundle has install path (`tool skill install [<host>]`) | no skill bundle (AGENTS.md or SKILL.md) found and --help declares no `skill` subcommand |
| N/A | `skill install --all` for multi-runtime install | no skill bundle (AGENTS.md or SKILL.md) found and --help declares no `skill` subcommand |
| N/A | `skill update` / `skill upgrade` for bundle refresh | no skill bundle (AGENTS.md or SKILL.md) found and --help declares no `skill` subcommand |
Reproduce this scorecard for xai-grok-build locally and inspect the failing audits:
anc audit --command grok --output jsonInstall anc first if you don't have it. Add --output json to get the same JSON shape committed under scorecards/.