Skip to content

Garry's Opinionated OpenClaw/Hermes Agent Brain

TypeScript
42.4 KB
Raw Blame
{
  "golden": [
    {
      "argv": "mcp",
      "exitCode": 0,
      "stderr": [],
      "stdout": [
        "gbrain mcp — provision access on the host; install it inside the harness",
        "",
        "gbrain mcp grant NAME --harness ID --profile PROFILE --source SOURCE --url URL --credentials-out FILE",
        "gbrain mcp grant NAME --client ID --if-version N --profile PROFILE --url URL --harness ID --dry-run",
        "gbrain mcp verify --client ID --harness ID --url URL --credentials-file FILE [--delegate]",
        "gbrain mcp admin --help",
        "gbrain mcp adapters | profiles",
        "gbrain mcp expose [--port N] [--funnel] [--surface verbs|starter|full] [--enable-dcr] [--no-tailscale] [--no-service] [--no-install] [--force] [--dry-run] [--yes] [--json]",
        "gbrain mcp expose --status [--json]",
        "gbrain mcp expose --remove [--yes] [--json]",
        "",
        "expose publishes the gbrain HTTP MCP server on your Tailscale tailnet (--funnel: public, for cloud agents)",
        "and keeps it running as a user service. Engine-free. See: gbrain mcp expose --help",
        "",
        "--profile defaults to memory-writer for new clients; omitted profiles preserve existing grants.",
        "New connections follow this brain's published skills within their approved read sources.",
        "--skills follow|memory-only         Follow shared skills (new default) or use memory only; existing grants are unchanged when omitted",
        "Following never grants skill editing, script execution, additional tools, or paid calls.",
        "Delegation requires --bound-tools T1,T2.",
        "--federated-read S1,S2              Explicit read sources",
        "--bound-slug-prefixes P1/,P2/       Direct write fence",
        "--delegated-slug-prefixes P1/,P2/   Delegated write fence",
        "--delegated-namespace job|prefixes  Default: job namespace",
        "--bound-max-concurrent N           Default for new clients: 1",
        "--budget-usd-per-day USD|unlimited Default for new clients: unlimited",
        "--token-ttl SECONDS                Override new access-token lifetime",
        "--admin-token-file FILE            Owner credential; otherwise GBRAIN_ADMIN_BOOTSTRAP_TOKEN",
        "--credentials-out FILE             Private credential handoff (required for creation)",
        "--credentials-file FILE            Existing private handoff; never printed",
        "--resume                           Recover the original private handoff for --client",
        "--timeout-ms N                     Bound verification (default: 30000)",
        "--delegate                         Execute a real worker check; may incur model charges",
        "--dry-run                          Preview grant without mutation or credentials",
        "--json                             Machine-readable redacted receipt",
        "",
        "Next, in the intended harness environment:",
        "gbrain connect URL --harness ID --credentials-file FILE --install",
        "Thin CLI adapters additionally require --root ABSOLUTE-PERSISTENT-DIRECTORY.",
        ""
      ]
    },
    {
      "argv": "import",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain import` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain import",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "access",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain access` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain access",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "export",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain export` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain export",
        "Why: export dumps the host brain to disk and has no MCP equivalent. Run `gbrain export` on the host, or read pages with the `get_page` MCP tool.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "files",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain files` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain files",
        "Why: `files list` and `files url` MCP ops are localOnly (paths live on the host filesystem). Use `gbrain files` on the host machine.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "embed",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain embed` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain embed",
        "Why: embed runs on the host. Run `gbrain embed` or `gbrain cycle` on the host machine.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "serve",
      "exitCode": 0,
      "stderr": [
        "[gbrain-bridge] relaying stdio to http://127.0.0.1:1/mcp surface=full (source: default)",
        "[gbrain-bridge] exit (stdin-end)"
      ],
      "stdout": []
    },
    {
      "argv": "sweep",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain sweep` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain sweep",
        "Why: sweep runs the serve-resident maintenance passes against the LOCAL engine. Run it on the host (the serve process also runs it automatically).",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "config",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain config` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain config",
        "Why: config reads/writes the host brain's config plane. Edit the host's .gbrain/config.json (file-plane keys) or run on the host with GBRAIN_HOME set. self_upgrade.* keys are machine-local and work here.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "pricing",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain pricing` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain pricing",
        "Why: pricing registers model prices in the host brain's config and is trusted-local only (a remote caller could void a cost cap). Ask the brain's operator to run `gbrain pricing set <model> ...` on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "migrate",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain migrate` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain migrate",
        "Why: migrate runs on the host's local engine. Run on the host machine.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "retrieval-upgrade",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain retrieval-upgrade` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain retrieval-upgrade",
        "Why: retrieval-upgrade (embedding migration) rebuilds the host brain's schema + re-embeds. Run on the host machine.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "eval",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain eval` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain eval",
        "Why: `eval` export/prune/replay touch the local engine and have no MCP equivalents. Run `gbrain eval` on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "jobs",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain jobs` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain jobs",
        "Why: `jobs list`, `jobs get <id>`, and `jobs stats` are thin-client routable; this subcommand runs against the host queue. Use the submit_job / list_jobs / get_job / get_job_stats MCP tools from your agent, or run on the host with GBRAIN_HOME set.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "agent",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain agent` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain agent",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "book-mirror",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain book-mirror` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain book-mirror",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "sync",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain sync` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain sync",
        "Why: sync runs on the host. Use the dedicated `sync_brain` MCP operation, or run `gbrain sync` on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "extract",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain extract` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain extract",
        "Why: extract runs on the host. Run `gbrain extract` or `gbrain cycle` on the host machine.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "extract-conversation-facts",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain extract-conversation-facts` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain extract-conversation-facts",
        "Why: extract-conversation-facts runs on the host (requires local engine + chat gateway). Run on the host machine.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "enrich",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain enrich` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain enrich",
        "Why: enrich runs on the host (requires local engine + chat gateway for grounded synthesis). Run on the host machine.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "features",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain features` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain features",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "autopilot",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain autopilot` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain autopilot",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "graph-query",
      "exitCode": 1,
      "stderr": [],
      "stdout": [
        "Usage: gbrain graph-query <slug> [options]",
        "",
        "Traverse the link graph from a page. Returns an indented tree of edges.",
        "Per-edge type filter: traversal only follows matching links.",
        "",
        "Options:",
        "  --type <link_type>     Filter to one link type (attended, works_at, invested_in,",
        "                         founded, advises, mentions, source).",
        "  --depth <N>            Max traversal depth (default 5).",
        "  --direction <dir>      'out' (default), 'in', or 'both'.",
        "  --source <id>          Scope the walk to this source. Defaults to the",
        "                         resolved source (GBRAIN_SOURCE, .gbrain-source,",
        "                         path match, brain default); __all__ spans every",
        "                         source. An unknown source is a hard error. Local",
        "                         installs only: a thin client rejects it (the server",
        "                         scopes the walk to your grant).",
        "  --hop <type>:<toward>  One hop of a typed chain; repeat for 2-3 hops. <toward>",
        "                         is 'object' (subject -> object, e.g. investor -> company)",
        "                         or 'subject' (object -> subject, e.g. company -> founder).",
        "                         Follows the relation's meaning, whichever page the edge",
        "                         was written on. Prints answers with the edges that prove",
        "                         each one. Cannot be combined with --type, --depth or",
        "                         --direction. MCP: traverse_graph {\"hops\":[{\"link_type\":",
        "                         \"invested_in\",\"toward\":\"object\"}, ...]}. Chain link",
        "                         types: founded, invested_in, led_round, advises, works_at, attended, yc_partner.",
        "                         Guide: docs/guides/multi-hop.md",
        "  --include-foreign      Include edges to pages in other sources (v0.37.7.0).",
        "                         Off by default; the walk stays inside the resolved",
        "                         source, and a footer reports the count of",
        "                         foreign-source edges hidden so users discover they exist.",
        "                         Not forwarded on a thin client.",
        "  -h, --help             Show this message.",
        "",
        "Examples:",
        "  gbrain graph-query people/alice --type attended --depth 2",
        "    -> who attended meetings with Alice (multi-hop)",
        "  gbrain graph-query companies/acme --type works_at --direction in",
        "    -> who works at Acme",
        "  gbrain graph-query people/bob --depth 1",
        "    -> Bob's direct connections",
        "  gbrain graph-query people/bob --include-foreign",
        "    -> include edges to pages in other sources",
        "  gbrain graph-query people/alice --hop invested_in:object --hop founded:subject",
        "    -> founders of the companies Alice invested in, with evidence",
        ""
      ]
    },
    {
      "argv": "reconcile-links",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain reconcile-links` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain reconcile-links",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "repair",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain repair` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain repair",
        "Why: repair runs on the brain host (it publishes coordinated page writes against the local engine). Run `gbrain repair` on the brain host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "diag",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain diag` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain diag",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "orphans",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain orphans` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain orphans",
        "Why: orphans needs the host's brain. Run on the host or use the `find_orphans` MCP tool from your agent.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "maintain",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain maintain` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain maintain",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "reindex",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain reindex` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain reindex",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "salience",
      "exitCode": 1,
      "stderr": [
        "Error [unavailable]: OAuth discovery failed: OAuth discovery network error: ECONNREFUSED: Unable to connect. Is the computer able to access the url?",
        "Fix: gbrain remote doctor",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#unavailable"
      ],
      "stdout": []
    },
    {
      "argv": "anomalies",
      "exitCode": 1,
      "stderr": [
        "[gbrain] reading anomalous cohorts; large brains can take minutes. Progress prints here every 10 s; the result prints on stdout when it is ready.",
        "Error [unavailable]: OAuth discovery failed: OAuth discovery network error: ECONNREFUSED: Unable to connect. Is the computer able to access the url?",
        "Fix: gbrain remote doctor",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#unavailable"
      ],
      "stdout": []
    },
    {
      "argv": "feedback",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain feedback` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain feedback",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "status",
      "exitCode": 0,
      "stderr": [
        "[gbrain] reading the brain status snapshot; large brains can take minutes. Progress prints here every 10 s; the result prints on stdout when it is ready."
      ],
      "stdout": [
        "",
        "GBrain Status",
        "=============",
        "Mode: thin-client  ·  v<version>  ·  <ts>",
        "",
        "Locks:",
        "  local-only — N/A on remote brain",
        "",
        "Autopilot:",
        "  local-only — N/A on remote brain",
        "",
        "Warnings:",
        "  ! remote snapshot failed: OAuth discovery failed: OAuth discovery network error: ECONNREFUSED: Unable to connect. Is the computer able to access the url?"
      ]
    },
    {
      "argv": "advisor",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain advisor` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain advisor",
        "Why: advisor ranks next actions from the host brain. Use the `advisor` MCP tool from your agent (when the host publishes it), or run `gbrain advisor` on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "conversation-parser",
      "exitCode": 0,
      "stderr": [],
      "stdout": [
        "Usage: gbrain conversation-parser <subcommand> [options]",
        "",
        "Subcommands:",
        "  scan <slug>          Dry-run the parser on a page; report pattern hit.",
        "  list-builtins        Print the built-in pattern registry.",
        "  validate <file>      Validate a user-declared simple_pattern JSON spec.",
        "                       (v1: simple_pattern compilation not yet implemented;",
        "                        emits \"TODO v0.42+\" notice.)",
        "",
        "Examples:",
        "  gbrain conversation-parser scan conversations/imessage/alice-example",
        "  gbrain conversation-parser list-builtins --json",
        "  gbrain conversation-parser validate ./my-pattern.json",
        "",
        "Global flags:",
        "  --json               Emit JSON envelope on stdout (where applicable).",
        "  --help, -h           Print this help."
      ]
    },
    {
      "argv": "edges-backfill",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain edges-backfill` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain edges-backfill",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "whoknows",
      "exitCode": 2,
      "stderr": [
        "gbrain whoknows: topic argument required",
        "Usage: gbrain whoknows <topic> [options]",
        "",
        "Ask your brain who knows about a topic. Returns ranked person/company",
        "pages by expertise depth, relationship recency, and salience.",
        "",
        "Options:",
        "  --limit N           Max results (default 5)",
        "  --explain           Show the ranking factor breakdown per result",
        "  --json              JSON output for agents",
        "  --help, -h          Show this help",
        "",
        "Examples:",
        "  gbrain whoknows \"lab automation\"",
        "  gbrain whoknows fintech compliance --explain",
        "  gbrain whoknows \"ai agents\" --limit 10 --json",
        ""
      ],
      "stdout": []
    },
    {
      "argv": "brainstorm",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain brainstorm` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain brainstorm",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "lsd",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain lsd` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain lsd",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "skillopt",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain skillopt` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain skillopt",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "calibration",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain calibration` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain calibration",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "transcripts",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain transcripts` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain transcripts",
        "Why: transcripts is server-private (raw chat exports stay on the host). Read transcripts on the host machine.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "models",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain models` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain models",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "takes",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain takes` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain takes",
        "Why: takes list/search/scorecard/calibration + add/update/resolve/supersede route to the brain host automatically (takes_* MCP ops). This subcommand (extract/revisit) is host-bound: run it on the host machine.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "onboard",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain onboard` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain onboard",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "founder",
      "exitCode": 1,
      "stderr": [
        "Unknown founder subcommand: (none). Did you mean \"founder scorecard\"?",
        "",
        "Usage: gbrain founder scorecard <entity-slug> [options]",
        "",
        "Rolls up the entity's typed-claim trajectory + resolved-take outcomes into",
        "the four founder-evaluation metrics: claim accuracy, consistency, growth",
        "trajectory, red flags.",
        "",
        "Options:",
        "  --since YYYY-MM-DD  Lower bound on valid_from / take created_at (default: 1 year ago)",
        "  --until YYYY-MM-DD  Upper bound (default: today)",
        "  --json              JSON output (stable schema_version: 1)",
        "  --help, -h          Show this help",
        "",
        "Examples:",
        "  gbrain founder scorecard companies/acme-example",
        "  gbrain founder scorecard companies/acme-example --since 2025-05-17 --json",
        ""
      ],
      "stdout": []
    },
    {
      "argv": "think",
      "exitCode": 0,
      "stderr": [],
      "stdout": [
        "Usage: gbrain think \"<question>\" [options]",
        "",
        "Options:",
        "  --anchor <slug>          Pull the entity subgraph around this slug",
        "  --rounds N               Multi-pass synthesis (default 1; gap-driven loop ships in v0.29)",
        "  --save                   Persist a synthesis page under synthesis/<slug>-<date>.md",
        "  --take                   Append a take row to the anchor page (requires --anchor)",
        "  --model <name>           Override the model: provider:model (preferred) or",
        "                           provider/model or a bare alias. An explicit --model that",
        "                           can't be resolved is a hard error (exit 1) — never a",
        "                           silent no-LLM degrade. On a thin client the remote",
        "                           brain answers with its own think model unless its",
        "                           host allows this one (think.remote_models).",
        "  --source <id>            Scope evidence gathering to this source. An unknown",
        "                           or archived source is a hard error (exit 1).",
        "  --since YYYY-MM-DD       Start of temporal window",
        "  --until YYYY-MM-DD       End of temporal window",
        "  --reference-date YYYY-MM-DD",
        "                           Date the question's relative time words (\"last month\")",
        "                           resolve against (default: today in brain.timezone)",
        "  --with-calibration       Inject the active calibration profile (anti-bias rewrite)",
        "  --calibration-holder <h> Holder whose calibration profile to use (default: self)",
        "  --json                   Output as JSON",
        "  --help                   Show this help",
        "",
        "Without --save, the synthesis is printed to stdout and discarded. With --save,",
        "the synthesis page is persisted AND printed. If --save is given but no synthesis",
        "was produced (no LLM available, or empty result), nothing is saved and the command",
        "exits non-zero.",
        "",
        "Dates: the reader sees today's date in brain.timezone and each page's content",
        "date; see docs/guides/time-aware-recall.md.",
        "",
        "Set ANTHROPIC_API_KEY (or run: gbrain config set anthropic_api_key ...) to run",
        "real synthesis. Without it AND without --save, the gather phase still runs and",
        "prints what would have been the input (exit 0).",
        ""
      ]
    },
    {
      "argv": "recall",
      "exitCode": 1,
      "stderr": [
        "Error [unavailable]: OAuth discovery failed: OAuth discovery network error: ECONNREFUSED: Unable to connect. Is the computer able to access the url?",
        "Fix: gbrain remote doctor",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#unavailable"
      ],
      "stdout": []
    },
    {
      "argv": "notability-eval",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain notability-eval` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain notability-eval",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "sources",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain sources` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain sources",
        "Why: sources commands manage local DB + config rows. Per-subcommand thin-client routing lands in v0.31.x. For now: use `sources_list` / `sources_status` MCP tools, or run on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "waiting",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain waiting` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain waiting",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "loops",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain loops` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain loops",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "questions",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain questions` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain questions",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "edge-proposals",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain edge-proposals` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain edge-proposals",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "memory-proposals",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain memory-proposals` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain memory-proposals",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "core",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain core` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain core",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "connectors",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain connectors` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain connectors",
        "Why: connectors manage provider session credentials in ~/.gbrain/connectors and sync your chat history on the host. Credentials never cross the wire — run on the host machine.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "wakeshift",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain wakeshift` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain wakeshift",
        "Why: `wakeshift watch job|write|cron` and `wakeshift accept-terms` route to the brain host automatically (watch_job / watch_write_request / watch_cron / accept_wakeshift_terms MCP tools). enable, disable, status and test read or change the brain host's Wakeshift key and settings: run those on the brain host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "pages",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain pages` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain pages",
        "Why: `pages purge-deleted` is admin+localOnly (hard-deletes from the local DB). Run on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "quarantine",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain quarantine` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain quarantine",
        "Why: `quarantine list` routes to the brain host automatically (quarantine_list MCP op). scan/clear are host-bound (bulk re-import; the clear trust decision) — run those on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "beliefs",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain beliefs` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain beliefs",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "guests",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain guests` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain guests",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "trust",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain trust` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain trust",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "watch",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain watch` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain watch",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "storage",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain storage` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain storage",
        "Why: storage operates on the local repo on disk. Run on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "code-def",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain code-def` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain code-def",
        "Why: `code-def` needs symbol-aware lookup that has no MCP op yet. Run on the host or use `search` from your agent with a symbol-shaped query.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "code-refs",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain code-refs` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain code-refs",
        "Why: `code-refs` has no MCP op yet. Run on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "reindex-code",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain reindex-code` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain reindex-code",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "reindex-search-vector",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain reindex-search-vector` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain reindex-search-vector",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "reindex-frontmatter",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain reindex-frontmatter` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain reindex-frontmatter",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "backfill",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain backfill` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain backfill",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "code-callers",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain code-callers` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain code-callers",
        "Why: `code-callers` has no MCP op yet. Run on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "code-callees",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain code-callees` is not routable (thin-client of http://127.0.0.1:1/mcp); run it on the brain host.",
        "Fix: gbrain code-callees",
        "Why: `code-callees` has no MCP op yet. Run on the host.",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    },
    {
      "argv": "repos",
      "exitCode": 1,
      "stderr": [
        "Error [requires_local_engine]: `gbrain repos` requires a local engine. This install is a thin client of http://127.0.0.1:1/mcp.",
        "Fix: gbrain repos",
        "Docs: https://github.com/garrytan/gbrain/blob/master/docs/guides/error-codes.md#requires_local_engine"
      ],
      "stdout": []
    }
  ],
  "normalizer": "cli-thin-client-v1"
}