memgineering 0.15.1 → 0.17.0

package/CHANGELOG.md

@@ -11,6 +11,52 @@ language the reader wants. The bilingual rule the monorepo applies to
 
 ## [Unreleased]
 
+## [0.17.0] — 2026-08-30
+
+### Added
+
+- **One sign-in, both halves.** `memgineering login` now links your registry
+  account automatically — same Google account, no extra step — and stores a
+  registry session next to the brain token. `memgineering whoami` shows both
+  identities (brain name and registry username); `memgineering logout` ends
+  both. If the registry half is ever unreachable during login, the brain
+  sign-in still completes and whoami says how to link later.
+- **Approving a login is one click now.** The browser approval screen keys off
+  your memgineering.com sign-in: signed in, you see who is asking and press
+  승인 — no separate Google round trip. Signed out, the same page signs you in
+  first (Google in a popup) and then asks. Consent is requested exactly once,
+  when an account is first created.
+
+### Agent guidance
+
+- `memgineering-setup` explains the unified sign-in: one login carries brain
+  and registry credentials together, and `whoami` reports both. Restart agent
+  sessions after upgrading so the refreshed guidance loads.
+
+## [0.16.0] — 2026-08-30
+
+### Added
+
+- **Opt-in anonymized error reporting.** When the CLI itself breaks — the
+  server unreachable, the server's own 500, or an unhandled crash — it can
+  send an anonymous report: error kind, CLI/Node version, platform, and a
+  sanitized message excerpt (paths, tokens, and identifiers stripped;
+  200-character cap). Never note content, never commands, never identity —
+  the server table has no columns identity could land in, and expected
+  refusals (like `no vault is linked`) are never reported. Off by default:
+  `memgineering setup` asks once in a terminal, `--error-telemetry on|off`
+  answers by flag, and `MEMGINEERING_NO_TELEMETRY=1` overrides everything.
+  Reports are deleted after 90 days. The privacy policy
+  (api.memgineering.com/privacy, effective 2026-08-30) documents all of it.
+
+### Agent guidance
+
+- `memgineering-setup` explains the telemetry question and both
+  off-switches, so an agent running setup for someone can relay the decision
+  instead of guessing it. Setup never prompts without a terminal — an
+  unanswered question stays `undecided`, which behaves as off. Restart your
+  agent session after upgrading so the refreshed skill is what gets read.
+
 ## [0.15.1] — 2026-08-29
 
 ### Fixed

package/assets/MEMGINEERING.md

@@ -2,7 +2,7 @@
 name: memgineering
 description: Use whenever the user refers to something they told you before, asks what was decided, tells you something worth keeping, or settles something that should hold next time. The memory lives in their own folder and outlives this session; check it before answering from guesswork, and write to it when you learn something durable.
 type: skill
-version: 0.15.1
+version: 0.17.0
 ---
 
 # memgineering

package/assets/memgineering-setup/SKILL.md

@@ -33,7 +33,8 @@ that can read it.
 memgineering setup --agent \
   --tools claude,codex \        # default: everything detected
   --auto-update on \            # default: on, and recommended
-  --hook on                     # session-start summary + rules before an edit
+  --hook on \                   # session-start summary + rules before an edit
+  --error-telemetry on          # anonymized crash reports; unset = undecided (off)
 ```
 
 `--dry-run` reports what would change without changing it.
@@ -44,6 +45,18 @@ opening nothing. Show the user the code — they approve it in their browser whi
 you keep working, and the next command picks the token up. Nothing exists on the
 server until they approve.
 
+**One sign-in covers both halves (0.17.0+).** The approval also links the
+user's registry account (same Google account, automatic), so the claim stores a
+registry session next to the brain token. `memgineering whoami` reports both —
+the brain identity and a `registry:` line with the username. If the registry
+line says "not linked", the link simply did not happen at sign-in time (older
+server, registry briefly down); a fresh `logout` + `login` links it. Never
+treat a missing registry line as a failed brain sign-in — the brain token
+stands on its own. On the approval page itself, a user already signed in to
+memgineering.com approves with one click; a signed-out user signs in right
+there first (Google in a popup). Consent is asked exactly once, when an
+account is first created.
+
 `--no-login` is how someone says "this machine only". Offer it, do not assume it:
 a brain that lives in a folder here cannot follow them to their phone or their
 next laptop, and that is discovered weeks later, by which time the notes are in
@@ -60,6 +73,17 @@ terminal at all.
 itself; `--hook on` means note summaries and their standing rules reach an agent
 unasked. Recommend both, but they are theirs to decide.
 
+**Error telemetry is theirs to decide too, and defaults to off.** With
+`--error-telemetry on`, the CLI sends an anonymous report when it ITSELF
+breaks: error kind, CLI/Node version, platform, and a sanitized message
+excerpt — never note content, commands, or anything naming them, and the
+server has no columns identity could land in. Unset leaves it `undecided`
+(behaves as off) so a person can be asked later; a terminal `setup` asks
+once; without a terminal it never prompts. Two off-switches at any time:
+`memgineering setup --error-telemetry off`, or `MEMGINEERING_NO_TELEMETRY=1`
+in the environment. Deleted server-side after 90 days; the full contract is
+at api.memgineering.com/privacy.
+
 **`--hook off` removes them**, it does not merely decline to add them, so it is
 the answer when someone already has the callbacks and has decided they do not
 want them. Only groups memgineering wrote are taken out; anything else in that