pp-v0

<!-- GENERATED FILE — DO NOT EDIT. This file is a verbatim mirror of library/ai/v0/SKILL.md, regenerated post-merge by tools/generate-skills/. Hand-edits here are silently overwritten on the next regen. Edit the library/ source instead. See the repository agent guide, section "Generated artifacts: registry.json, cli-skills/". --> # v0 — Printing Press CLI ## Prerequisites: Install the CLI This skill drives the `v0-pp-cli` binary. **You must verify the CLI is installed before invoking any command from this skill.** If it is missing, install it first: 1. Install via the Printing Press installer. It defaults binaries to `$HOME/.local/bin` on macOS/Linux and `%LOCALAPPDATA%\Programs\PrintingPress\bin` on Windows: ```bash npx -y @mvanhorn/printing-press-library install v0 --cli-only ``` 2. Verify: `v0-pp-cli --version` 3. Ensure the reported install directory is on `$PATH` for the agent/runtime that will invoke this skill. If the `npx` install fails (no Node, offline, etc.), fall back to a direct Go install (requires Go 1.26.5 or newer). This installs into `$GOPATH/bin` (default `$HOME/go/bin`), so add that directory to `$PATH` instead: ```bash go install github.com/mvanhorn/printing-press-library/library/ai/v0/cmd/v0-pp-cli@latest ``` If `--version` reports "command not found" after install, the runtime cannot see the binary directory on `$PATH`. Do not proceed with skill commands until verification succeeds. The v0 API v2 CLI with offline search, streaming capture, and credit-spend analytics — no other tool tracks where your v0 credits go. ## When to Use This CLI Use v0-pp-cli to generate and iterate on web apps via the v0 API: creating chats from prompts, sending follow-up messages, fetching generated files and previews, deploying to Vercel, managing MCP servers and webhooks, and tracking credit spend. ## Anti-triggers Do not use this CLI for: - Do not use this CLI for editing files inside a generated app's sandbox directly; fetch files and edit locally instead. - Do not use this CLI to manage Vercel infrastructure outside v0 chats; that is Vercel's own API surface. - Do not use this CLI as a general-purpose coding agent; it generates apps through v0, not in the current directory. ## Unique Capabilities These capabilities aren't available in any other tool for this API. ### Cost intelligence - **`spend`** — Aggregate v0 credit cost and token usage from the synced message mirror, grouped by chat, day, or model. _Use this when an agent needs to know where v0 credits went, which chats are expensive, or a daily burn rate._ ```bash v0-pp-cli spend --since 7d --by chat --json ``` ### Streaming - **`chats stream`** — Create a chat and stream the SSE response live, rendering each event and recording model attribution for spend analytics. _Use this when an agent needs to see generation progress live or capture the event stream for later analysis._ ```bash v0-pp-cli chats stream "Create a kanban dashboard" --model v0-pro ``` - **`messages tail`** — Poll a chat until the newest assistant message finishes, with --follow for continuous watching. _Use this when an agent kicked off an async generation and needs to wait for it deterministically._ ```bash v0-pp-cli messages tail ft7dqhYEX8n --interval 3s --timeout 10m ``` ### Files - **`chats files`** — Render a chat's generated source files as an indented directory tree. _Use this when an agent needs to understand a generated app's layout before editing._ ```bash v0-pp-cli chats files ft7dqhYEX8n --tree ``` - **`chats preview`** — Print only the live preview URL for a chat, ready for embedding or scripting. _Use this when an agent needs the preview URL for CI, screenshots, or quick checks._ ```bash v0-pp-cli chats preview ft7dqhYEX8n --url ``` ### Offline - **`search`** — Full-text search over synced chats and messages without hitting the API. _Use this when an agent needs to find a past chat or message without paging the API._ ```bash v0-pp-cli search "kanban" --json ``` - **`sync`** — Cursor-paginated sync of chats and messages into a local SQLite mirror for offline search and spend analytics. _Run once before offline search or spend commands._ ```bash v0-pp-cli sync --resources chats,messages ``` ### Ops - **`doctor`** — Validate V0_API_KEY and API reachability with a live check. _Use this when a script 401s and you need to know why._ ```bash v0-pp-cli doctor ``` ## Command Reference **chats** — Generate and manage apps from prompts - `v0-pp-cli chats connect-status` — Get the setup status of a Vercel Connect integration - `v0-pp-cli chats create` — Create a chat from a prompt (blocks until the model response is complete) - `v0-pp-cli chats create-async` — Create a chat asynchronously; poll the returned messageId for completion - `v0-pp-cli chats create-from-files` — Create a chat from source files - `v0-pp-cli chats create-from-repo` — Create a chat from a GitHub repository - `v0-pp-cli chats create-from-zip` — Create a chat from a ZIP archive URL - `v0-pp-cli chats create-stream` — Create a chat and stream the model response as Server-Sent Events - `v0-pp-cli chats create-vercel-project` — Create a Vercel project for a chat - `v0-pp-cli chats delete` — Delete a chat permanently - `v0-pp-cli chats deploy` — Deploy a chat to Vercel - `v0-pp-cli chats download-files` — Download all chat files as an archive - `v0-pp-cli chats duplicate` — Duplicate a chat - `v0-pp-cli chats files` — Get all source files in a chat - `v0-pp-cli chats get` — Get a chat by ID - `v0-pp-cli chats list` — List chats accessible to the authenticated user - `v0-pp-cli chats preview` — Get the preview URL and short-lived access token for a chat - `v0-pp-cli chats restore-message` — Restore files from a previous assistant message - `v0-pp-cli chats resume-stream` — Reconnect to an active chat stream - `v0-pp-cli chats update` — Update a chat title, privacy, or metadata - `v0-pp-cli chats update-files` — Create, update, or delete files in a chat **hooks** — Manage webhooks that listen for chat and message events - `v0-pp-cli hooks create` — Create a webhook subscribed to chat and message events - `v0-pp-cli hooks delete` — Delete a webhook - `v0-pp-cli hooks get` — Get a webhook by ID - `v0-pp-cli hooks list` — List all webhooks in the workspace - `v0-pp-cli hooks update` — Update a webhook configuration **mcp-servers** — Manage MCP server connections for chats (max 10 per user) - `v0-pp-cli mcp-servers create` — Register a new MCP server - `v0-pp-cli mcp-servers delete` — Delete an MCP server - `v0-pp-cli mcp-servers get` — Get an MCP server by ID - `v0-pp-cli mcp-servers list` — List MCP servers configured for the account - `v0-pp-cli mcp-servers update` — Update an MCP server configuration **messages** — Send, list, and manage chat messages - `v0-pp-cli messages get` — Get a single message - `v0-pp-cli messages list` — List messages in a chat, newest first - `v0-pp-cli messages resolve-task` — Resolve a chat blocked waiting for user input - `v0-pp-cli messages resolve-task-async` — Resolve a blocked chat asynchronously - `v0-pp-cli messages resolve-task-stream` — Resolve a blocked chat and stream the response - `v0-pp-cli messages send` — Send a message and wait for the model response - `v0-pp-cli messages send-async` — Send a message asynchronously; poll the returned messageId for completion - `v0-pp-cli messages send-stream` — Send a message and stream the response as Server-Sent Events - `v0-pp-cli messages stop` — Stop an in-flight message generation **settings** — Manage workspace settings - `v0-pp-cli settings preview-hosts` — Get trusted hostname patterns allowed to embed previews - `v0-pp-cli settings set-preview-hosts` — Set the complete list of trusted preview hosts ### Finding the right command When you know what you want to do but not which command does it, ask the CLI directly: ```bash v0-pp-cli which "<capability in your own words>" ``` `which` resolves a natural-language capability query to the best matching command from this CLI's curated feature index. Exit code `0` means at least one match; exit code `2` means no confident match — fall back to `--help` or use a narrower query. ## Recipes ### Create a chat and wait for it ```bash v0-pp-cli chats create --message "A minimal landing page" --title Landing --privacy private --json ``` Blocking create returns the chat plus token/credit usage once the model finishes. ### Stream a generation and record the model ```bash v0-pp-cli chats stream "A pricing page" --model v0-pro --privacy private ``` Streams SSE events live and records model attribution so spend --by model works. ### Send a follow-up asynchronously and tail it ```bash v0-pp-cli messages send-async ft7dqhYEX8n --message "Add dark mode" --json ``` Async send returns a messageId; poll it with messages tail until finishReason is set. ### Inspect generated files as a tree ```bash v0-pp-cli chats files ft7dqhYEX8n --tree ``` Renders the generated app's file layout without downloading the archive. ### Track weekly credit spend by chat ```bash v0-pp-cli spend --since 7d --by chat --json ``` After sync, aggregates usage.creditsCost and tokens from the local mirror. ### Embed a live preview ```bash v0-pp-cli chats preview ft7dqhYEX8n --url ``` Prints just the preview URL for embedding or scripting. ## Auth Setup Requires a v0 API key (create one at https://v0.app/settings/keys). Set V0_API_KEY in your environment, or use auth set-token. The CLI sends Authorization: Bearer <key>. Run `v0-pp-cli doctor` to verify setup. ## Agent Mode Add `--agent` to any command. Expands to: `--json --compact --no-input --no-color --yes`. - **Pipeable** — JSON on stdout, errors on stderr - **Filterable** — `--select` keeps a subset of fields. Dotted paths descend into nested structures; arrays traverse element-wise. Critical for keeping context small on verbose APIs: ```bash v0-pp-cli chats list --agent --select id,name,status ``` - **Previewable** — `--dry-run` shows the request without sending - **Offline-friendly** — sync/search commands can use the local SQLite store when available - **Non-interactive** — never prompts, every input is a flag - **Explicit retries** — use `--idempotent` only when an already-existing create should count as success, and use `--ignore-missing` only when a missing delete target should count as success ### Response envelope Commands that read from the local store or the API wrap output in a provenance envelope: ```json { "meta": {"source": "live" | "local", "synced_at": "...", "reason": "..."}, "results": <data> } ``` Parse `.results` for data and `.meta.source` to know whether it's live or local. A human-readable `N results (live)` summary is printed to stderr only when stdout is a terminal AND no machine-format flag (`--json`, `--csv`, `--compact`, `--quiet`, `--plain`, `--select`) is set — piped/agent consumers and explicit-format runs get pure JSON on stdout. ## Paths and state Agents should treat the CLI's path resolver as part of the runtime contract: - Use `--home <dir>` for one invocation, or set `V0_HOME=<dir>` to relocate all four path kinds under one root. - Use per-kind env vars only when a specific kind must diverge: `V0_CONFIG_DIR`, `V0_DATA_DIR`, `V0_STATE_DIR`, `V0_CACHE_DIR`. - Resolution order is per-kind env var, `--home`, `V0_HOME`, XDG (`XDG_CONFIG_HOME`, `XDG_DATA_HOME`, `XDG_STATE_HOME`, `XDG_CACHE_HOME`), then platform defaults. - `config` contains settings like `config.toml` and profiles. `data` contains `credentials.toml`, `data.db`, cookies, and auth sidecars. `state` contains persisted queries, jobs, and `teach.log`. `cache` contains regenerable HTTP/cache files. - Stored secrets live in `credentials.toml` under the data dir. Existing legacy `config.toml` secrets are read for compatibility and leave `config.toml` on the first auth write. - Run `v0-pp-cli doctor --fail-on warn` to surface path and credential-location warnings. `agent-context` exposes a schema v4 `paths` block for agents that need the resolved dirs. - For MCP, pass relocation through the MCP host config. The MCP binary does not inherit CLI flags: ```json { "mcpServers": { "v0": { "command": "v0-pp-mcp", "env": { "V0_HOME": "/srv/v0" } } } } ``` Fleet precedence: an inherited per-kind env var overrides an explicit `--home` for that kind. Use `V0_HOME` or per-kind vars as durable fleet levers, and use `--home` only for a single invocation. Relocation is not reversible by unsetting env vars; move files manually before clearing `V0_HOME`, or `doctor` will not find credentials left under the former root. ## Automatic learning This CLI ships a self-capturing learning loop. The CLI does its own bookkeeping: every invocation is journaled locally, a failed flag followed by a corrected retry auto-derives a `flag_alias` candidate, and a `teach` on a query family without a playbook auto-synthesizes a `playbook_candidate` from the session's journal. Your job is judgment only: `recall` first, act on surfaced candidates, `teach` the final answer, `playbook amend` when you observe a correction. You never record failures by hand. ### Step 1: `recall` before any discovery Before list/search/drill commands on a new user question, run: ```bash v0-pp-cli recall "<user's question>" --agent ``` The response envelope: ```json { "query": "...", "normalized": "<normalized form>", "query_entities": ["..."], "found": true | false, "match_score": 0.0, "results": [ { "resource_id": "...", "resource_type": "...", "venue": "...", "confidence": 2, "entity_match": "exact|partial|unknown", "source": "taught|preseed|pattern", "warnings": ["..."] } ], "mismatches": [ /* only when --debug-mismatches */ ], "warnings": [ /* top-level */ ], "candidates": [ { "id": 12, "class": "flag_alias | playbook_candidate", "summary": "...", "sightings": 3, "last_seen": "...", "rationale": "...", "next_action": ["<trial command>", "v0-pp-cli learnings confirm 12"] } ], "playbook": { "query_family": "...", "playbook": { "steps": [ { "cmd": "<command with {slot} substitution>", "purpose": "..." } ], "entity_slots": ["$ENTITY"], "expected_tool_calls": 3 }, "slots_resolved": { "$ENTITY": { "token": "<live token>", "canonical": "<canonical>" } }, "notes": "<workarounds + gotchas for this query family>" }, "notes": "<duplicate surface for non-playbook callers>" } ``` Empty-store short-circuit: if the store has no learnings, playbooks, or candidates yet (recall finds nothing and `learnings list` and `learnings candidates` are both empty), skip recall for the rest of this session instead of taxing every query; resume recall-first once something has been taught. ### Step 2: decision tree Read `candidates`, `playbook`, `notes`, `results[0]`, and warnings in that order: