--- name: pp-google-voice description: "Read Google Voice through ordinary HTTP and recover evidence from a local message mirror. Trigger phrases: `check my Google Voice messages`, `sync Google Voice history`, `search my Voice messages`, `what Google Voice history is collected`, `use google-voice`, `run google-voice`." author: "mikemarqq" license: "Apache-2.0" argument-hint: " [args] | install cli|mcp" allowed-tools: "Read Bash" metadata: openclaw: requires: bins: - google-voice-pp-cli install: - kind: go bins: [google-voice-pp-cli] module: github.com/mvanhorn/printing-press-library/library/social-and-messaging/google-voice/cmd/google-voice-pp-cli --- # Google Voice — Printing Press CLI > **Pull-request builds:** before the catalog release is available, build and run both local executables following README.md. The standard catalog-install section applies after merge and release. Verify the executable before proceeding. This reader is verified on macOS. Voice session and mirror storage use Unix ownership and permission checks; Windows storage fails closed until directory ACL validation is implemented. ## Prerequisites: Install the CLI This skill drives the `google-voice-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 `$HOME/.local/bin` on macOS/Linux and `%LOCALAPPDATA%\Programs\PrintingPress\bin` on Windows: ```bash npx -y @mvanhorn/printing-press-library install google-voice --cli-only ``` 2. Verify: `google-voice-pp-cli --version` 3. Ensure the reported install directory is on `$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 1.26.10 or newer). This installs into `$GOPATH/bin` (default `$HOME/go/bin`), so add that directory to `$PATH` instead: ```bash go install github.com/mvanhorn/printing-press-library/library/social-and-messaging/google-voice/cmd/google-voice-pp-cli@latest ``` If `--version` reports "command not found" after install, the runtime cannot see the binary directory on `$PATH`. Do not proceed with skill commands until verification succeeds. Collect account-scoped message history with resumable pagination, then search it locally. Inspect surrounding conversation, observed activity and collection coverage before answering from that history. ## When to Use This CLI Use for read-only Google Voice account, conversation and message access and for deterministic queries over its local mirror. Check collection coverage when answering a question from missing or older history. Keep relationship judgment and cross-service person identity in separate tools. ## Anti-triggers Do not use this CLI for: - Do not use this CLI to send SMS/MMS, reply, call, mark read, archive or delete source conversations. - Do not use this CLI to merge identities or modify the existing Signal/Telegram relationship-memory database. - Do not use this CLI to install a background service or promise complete Google Voice history from one bounded sync. ## Unique Capabilities ### Recover message evidence - **`history query`** — Find messages using participant, dates, direction, literal text and attachment evidence together. _Use when a concrete question needs a bounded intersection of message facts._ ```bash google-voice-pp-cli history query --text appointment --direction inbound --limit 20 --agent ``` - **`history context`** — Retrieve neighboring messages around a known message, including collection boundaries. _Use after finding an exact message when its surrounding exchange matters._ ```bash google-voice-pp-cli history context --message-id sms-example --before 4 --after 4 --agent ``` - **`history activity`** — See observed inbound and outbound counts and latest message times per conversation. _Use to establish recent communication facts before a separate skill interprets them._ ```bash google-voice-pp-cli history activity --since 30d --limit 50 --agent ``` ### Understand collection state - **`mirror coverage`** — See freshness, unfinished pagination and collection failures before relying on a lookup. _Use before treating an empty search as evidence that no communication occurred._ ```bash google-voice-pp-cli mirror coverage --agent ``` - **`history arrivals`** — Inspect which records collection added, keeping old backfill distinct from recent communication. _Use to review collection growth and confirm repeated imports do not invent new arrivals._ ```bash google-voice-pp-cli history arrivals --limit 100 --agent ``` ## Command Reference **account** — Verify the imported Google Voice account - `google-voice-pp-cli account` — Read the Google Voice account bound to the private imported session **threads** — Read Google Voice conversation inventory and messages - `google-voice-pp-cli threads get` — Read one bounded message page without changing conversation read state - `google-voice-pp-cli threads list` — List a bounded conversation page with explicit continuation and folder evidence **history** - `google-voice-pp-cli history query` — Find messages using participant, dates, direction, literal text and attachment evidence together. - `google-voice-pp-cli history context` — Retrieve neighboring messages around a known message, including collection boundaries. - `google-voice-pp-cli history activity` — See observed inbound and outbound counts and latest message times per conversation. - `google-voice-pp-cli history arrivals` — Inspect which records collection added, keeping old backfill distinct from recent communication. **mirror** - `google-voice-pp-cli mirror coverage` — See freshness, unfinished pagination and collection failures before relying on a lookup. ### Finding the right command When you know what you want to do but not which command does it, ask the CLI directly: ```bash google-voice-pp-cli which "" ``` `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. `--json` (and other machine formats) keep that exit-2 contract and write `{"matches":[]}` on stdout so agents can inspect the envelope without treating a miss as success. ## Recipes ### Find inbound appointment messages ```bash google-voice-pp-cli history query --text appointment --direction inbound --limit 20 --agent --select messages.id,messages.thread_id,messages.timestamp,messages.text ``` Narrow literal evidence while preserving source identifiers. ### Inspect observed communication ```bash google-voice-pp-cli history activity --since 30d --limit 50 --agent ``` Return counts and last-observed source times without assigning social meaning. ### Check collection coverage ```bash google-voice-pp-cli mirror coverage --agent ``` Inspect freshness, inventory, older-history cursors and failed work. ### Inspect newly collected records ```bash google-voice-pp-cli history arrivals --limit 100 --agent ``` Separate first-seen collection time from original message time. ## Auth Setup Google Voice uses an existing signed-in Google web session, not a consumer API key. Import credentials from a private local HAR with auth import-har; the CLI keeps only allowlisted session material in a machine-local profile and computes SAPISIDHASH for each fixed-host read request. Expired sessions require a fresh private import; a running browser is not required for ordinary reads. Run `google-voice-pp-cli doctor` to verify setup. ## Agent Mode Add `--agent` to any command. Expands to: `--json --compact --no-input --no-color`. Global format flags share one contract on promoted, novel, sync, and `--deliver` paths: - `--json` — one JSON document on stdout (sync progress events go to stderr) - `--compact` — keep identity/status/timestamp fields; does not change the document vs stream shape - `--csv` / `--plain` — tabular rows (collection envelopes unwrap to the row array). `--csv` prefixes `'` on a cell or header key that starts with `=`, `+`, `-`, `@`, tab, CR, or LF (fullwidth `＝＋－＠` too) so a spreadsheet does not run it as a formula; a plain signed number such as `-5` or `+2.5e3` is unchanged. - `--quiet` — one identity value per row, no envelope - **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 google-voice-pp-cli account get --agent --select id,version ``` - **Previewable** — `--dry-run` shows the request without sending. A completed dry run exits 0 with `"success": true` and `"dry_run": true` and no HTTP `status`; validation failures still exit non-zero with `"success": false`. - **Offline-friendly** — sync/search commands can use the local SQLite store when available - **Non-interactive** — never prompts, every input is a flag - **Read-only** — do not use this CLI for create, update, delete, publish, comment, upvote, invite, order, send, or other mutating requests ### Response envelope With `--agent`, commands wrap their local or live output in a provenance envelope; `--json` alone preserves the command's documented raw JSON payload: ```json { "meta": {"source": "live" | "local", "synced_at": "...", "reason": "..."}, "results": } ``` For `--agent`, parse `.results` for data and `.meta.source` for its source. 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. ## Voice collection contract Use `mirror sync --max-pages 5` for bounded resumable collection, then inspect `mirror coverage`. Unknown folders, refresh needs and older-history cursors remain explicit; observed timestamp bounds never prove complete history. `history arrivals` uses collection time and first-seen source IDs; `history activity` uses original message time. Exact participant identifiers and separate thread IDs are preserved without person merging. `GOOGLE_VOICE_PROFILE` selects account-specific session and mirror paths; it is separate from the framework saved-run `--profile` flag. Voice session and SQLite paths are `voice-profiles//session.json` under config and `voice-profiles//messages.sqlite3` under data. The framework learning store is separate. Keep message bodies, HARs, credentials and exports local unless Mike explicitly requests a particular transmission. No scheduler or relationship-memory adapter is installed. ## Paths and state Agents should treat the CLI's path resolver as part of the runtime contract: - Use `--home ` for one invocation, or set `GOOGLE_VOICE_HOME=` to relocate all four path kinds under one root. - Use per-kind env vars only when a specific kind must diverge: `GOOGLE_VOICE_CONFIG_DIR`, `GOOGLE_VOICE_DATA_DIR`, `GOOGLE_VOICE_STATE_DIR`, `GOOGLE_VOICE_CACHE_DIR`. - Resolution order is per-kind env var, `--home`, `GOOGLE_VOICE_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. - Framework-managed credentials, when used, live in `credentials.toml` under data. Voice cookies use the separate private `session.json` described in Voice collection contract; importing a HAR does not migrate framework credentials. - Run `google-voice-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": { "google-voice": { "command": "google-voice-pp-mcp", "env": { "GOOGLE_VOICE_HOME": "/srv/google-voice" } } } } ``` Fleet precedence: an inherited per-kind env var overrides an explicit `--home` for that kind. Use `GOOGLE_VOICE_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 `GOOGLE_VOICE_HOME`, or `doctor` will not find credentials left under the former root. Local data is unencrypted and relies on OS account permissions plus disk encryption. Voice live reads bypass the HTTP response cache and do not automatically populate the message mirror; `mirror sync` is the explicit collection operation. `local-data purge` previews and, with `--yes`, erases framework-owned store/cache/journal data only; it retains Voice profile message mirrors, custom `--db` files and credentials, and its help states that boundary. `GOOGLE_VOICE_NO_LOCAL_DATA=1` disables framework response caching and blocks message collection through `mirror sync`; local queries can still read an existing mirror. `GOOGLE_VOICE_DATA_SOURCE` selects the read-mode default, and per-command source restrictions still apply. ## 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, pass the question as an argv or MCP tool argument to `recall --agent`. Do not interpolate user-controlled text into a shell command line. Quoted `recall ""` breaks on an apostrophe, which is ordinary English. A quoted heredoc breaks when a body line equals the delimiter, and that delimiter is published in these docs. Write the question with a non-shell file-writing tool, then read it back as data: ```bash # Write the question verbatim with your file-writing tool (no shell involved). # Command substitution on a file only ever yields data — the shell never # parses the file's bytes as syntax. QUERY=$(cat /path/to/question.txt) google-voice-pp-cli recall "$QUERY" --agent ``` Prefer MCP: pass the question as the tool's query argument. `"$QUERY"` after a file read is argv-safe; putting the question itself in the command text is not. The response envelope: ```json { "query": "...", "normalized": "", "query_entities": ["..."], "found": true | false, "match_score": 0.0, "results": [ { "resource_id": "...", "resource_type": "...", "venue": "...", "confidence": 2, "entity_match": "exact|partial|unknown", "source": "taught|preseed|pattern", "warnings": ["..."] } ], "mismatches": [ /* only when --debug-mismatches */ ], "warnings": [ /* top-level */ ], "candidates": [ { "id": 12, "class": "flag_alias | playbook_candidate", "summary": "...", "sightings": 3, "last_seen": "...", "rationale": "...", "next_action": ["", "google-voice-pp-cli learnings confirm 12"] } ], "playbook": { "query_family": "...", "playbook": { "steps": [ { "cmd": "", "purpose": "..." } ], "entity_slots": ["$ENTITY"], "expected_tool_calls": 3 }, "slots_resolved": { "$ENTITY": { "token": "", "canonical": "" } }, "notes": "" }, "notes": "" } ``` 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: ``` if Candidates present (warnings include "candidates_present"): -> candidates are try-then-confirm, never facts. Follow each candidate's two-step next_action verbatim: run the trial command first, then run `learnings confirm ` only after the trial verified the behavior. Reject a wrong candidate with `learnings reject `. -> NEVER re-teach something recall surfaced as a candidate; confirm or reject that candidate instead of teaching a duplicate. -> candidates ride alongside playbooks and resource hits, not instead of them; continue with the branches below after acting on them. if Playbook present: -> READ Playbook.notes verbatim FIRST (workarounds + gotchas the CLI surface doesn't expose) -> replay Playbook.steps in order, substituting Playbook.slots_resolved entries for the entity slot tokens. If a step's slot is unresolved, fall back to discovery for that step only. -> the Playbook's expected_tool_calls is a budget; if you find yourself running materially more, record the divergence via `google-voice-pp-cli playbook amend` at end-of-session. elif Notes present (no Playbook): -> read Notes verbatim before any discovery step; they carry known gotchas for this query family even when no structured choreography exists yet. elif Found AND Results[0].EntityMatch == "exact" AND Results[0].Confidence >= 2: -> skip discovery; fetch live data for Results[*].ResourceID in parallel elif Found AND Results[0].EntityMatch == "partial": -> candidate hint, NOT a hit; read the resource title to validate before trusting elif (any row in Mismatches[] when --debug-mismatches was passed): -> treat as cold start; the stored learning is for a different entity (different canonical resolved from query_entities) else: // Found == false, no playbook, no notes -> cold start; run discovery normally; teach the answer afterward (Step 4). If the family has no playbook yet, that teach auto-synthesizes a playbook candidate from this session's journal - you do not need to record one by hand. ``` Playbook and Notes are orthogonal to the per-resource path. A recall response can carry both a Playbook AND a `Results[]` hit - use both: the Playbook tells you which choreography to run; the resource hits short-circuit specific steps. Default to skipping `mismatches`; pass `--debug-mismatches` only when investigating cold-start surprises. Candidate judgment details: `learnings confirm ` prints the candidate's full payload before materializing it - check that the printed payload matches the behavior you verified. `learnings reject ` tombstones the derivation signature so the same candidate does not resurface. The envelope carries only the few candidates worth acting on now; `google-voice-pp-cli learnings candidates` lists the full op