← ClayCONTENT HISTORY

Update to Clay

Snapshot Nov 18, 2026 · 06:19 UTC · version 2.38.0

Collection source: downloaded plugin package.

WHAT CHANGED

First saved snapshot

No earlier snapshot is available to establish a change.

Save Clay and check future changes in Following.

Up to 3 plugins free · no card required.

Compare saved observations

Download comparison JSON
Full technical diff · 0 changed fields
Full snapshot data
{
  "description": "Clay CLI — the primary scripting surface (JSON output, typed errors). Use when running or discovering `clay` commands. For what Clay can do and which skill to use, read the `clay` skill when it is available; otherwise run `clay --help`. Use `clay --help` for command names, flags, JSON shape, and error codes.",
  "included_files": [],
  "name": "cli",
  "skill_md_contents": "---\nname: cli\ndescription: Clay CLI — the primary scripting surface (JSON output, typed errors). Use when running or discovering `clay` commands. For what Clay can do and which skill to use, read the `clay` skill when it is available; otherwise run `clay --help`. Use `clay --help` for command names, flags, JSON shape, and error codes.\n---\n\n# The `clay` CLI\n\nBefore using this skill, read the `clay` entry skill if you haven’t already this session, and follow its access guidance.\n\nThe `clay` CLI provides command-line access optimized for agents: JSON\noutput and typed error codes. For tool access or authentication problems, read the `clay` entry skill before continuing. The workspace is resolved from\nthe stored session — there is no workspace id to pass.\n\n## Use workspace context internally\n\nOnce the CLI knows which workspace a credential is for, every successful command's JSON\ncarries a `workspace: { id, name }` key naming the workspace it authenticated as. Use this\ninformation internally. Only mention authentication, the user, or the workspace when the\nuser asks or an actual account/workspace issue needs their attention. Running the first\ncommand, receiving identity in a result, or loading this skill later in a conversation\ndoes not call for an identity announcement. This applies to managed and standalone CLI\nsessions. Don't ask the user for workspace information merely because the key is absent — it is\nabsent when a command never had to resolve the workspace, and when its result is a top-level\nJSON array with nowhere to put it.\n\nA browser sign-in covers every workspace the user selects on the consent screen, and leaves\nthe first one selected active; `clay login --device` covers one workspace per run. Either way\nthe workspaces already signed in are kept, so `clay login` again adds more rather than\nreplacing them. `clay workspaces list` shows the ids and `clay workspaces switch <id>` moves\nbetween them, so switch explicitly to work in one the sign-in did not leave active. When a\npayload carries no workspace key and you need to know, `clay workspaces current` answers\ndirectly. Inside a managed Clay session none of this applies: the session runs in the one\nworkspace it was created in, and `clay login` / `clay logout` / `clay workspaces` are\nunavailable there. Never ask the user to sign in — but if they ask to work in a different\nworkspace, tell them to switch workspaces in Clay and start a new chat there, since this\nsession cannot move.\n\n## Discovering commands\n\nWhen a user asks what they can do with Clay, use the `clay` skill when it is available — that is the table of\ncontents for product surfaces. This skill is how to invoke the CLI.\n\n`clay --help` is the live list of top-level commands. The help text is a machine-readable\nspec: use it for command names, flags, JSON output shape, and error codes. Do not treat\nit as the answer to \"what can I do with Clay?\"\n\n```bash\nclay --help                 # top-level commands\nclay <group> --help         # a group's subcommands\nclay <group> <cmd> --help   # exact flags, JSON output shape, and error codes\n```\n\n## Prefer simple, single commands\n\nPrefer running one plain `clay` command at a time. Avoid redirecting or\nsubstituting (`&&`, `||`, `>`, `$(…)`, backticks, `$VAR`, etc.) unless it's genuinely\nnecessary — those forms fall through to a manual approval prompt, whereas a simple\n`clay <group> <cmd>` call is auto-approved.\n\nPiping (`|`) is fine when the other stages are common read-only helpers like `jq` that\ntransform stdin without opening files. Semicolon chaining (`;`) is narrower: each\nclause must be `clay`, or a literal `echo` / `printf` (not `cat` / `jq` / …). Anything\nelse falls through to a prompt.\n\n## When a command needs approval\n\nSome commands that delete, send or run things at scale can answer `approval_required`\ninstead of running. Nothing has happened yet: tell the user what the request would do (from\nthe error's `details`) and let them decide. `clay approvals approve <requestId>` runs it and\n`clay approvals deny <requestId>` drops it. Inside a managed Clay session the person decides in\nClay instead, and the `clay approvals` commands are unavailable.\n\nWhen `clay approvals get <requestId>` shows an `offeredRule`, `clay approvals approve <requestId>\n--always-allow` also saves it, so the user's later CLI calls for that action run without asking.\nOnly add `--always-allow` when the user asks for it. `clay approvals rules list` shows their saved\nrules and `clay approvals rules delete <ruleId>` removes one.\n\n## When to use the CLI vs the Public API\n\nUse the CLI for scripting and agent-driven tasks in a shell. To build a service, app,\nor integration that talks to Clay over HTTP, use the **Public API** (`public-api` skill).\n\nFull developer documentation (CLI reference, Public API reference, concepts, OpenAPI\nspec) lives at: https://developers.clay.com/llms.txt\n"
}

SHA-256 of public snapshot: 5b472b60a16e589c16bf56687fae70c7f48387e3cf7f049d9871c6bbe5b90713