Pulse is memory for AI tools. It gives Codex and Claude Code the knowledge they
need at the moment they need it, and stays silent when nothing relevant is
found. Version 0.8.2 can also remember emotions attached to a specific
moment. It does not turn repeated emotions into personality traits, save the
full conversation, or send Personal memory to a cloud server.
0.8.2 is the npm preview, not the stable default. Its release flow installs the exact archive on a clean Apple Silicon runner and makes a real BGE-M3 semantic query before publication. Use 0.7.2 for the stable installation. Version 0.8.1 remains preserved as evidence of the cold-search defect that 0.8.2 fixes.
Version 0.8.2 expands historical import, returns up to four short, distinct memories, keeps the embedder protocol intact after a cancelled caller, and makes Doctor prove a real semantic query. Owner-machine daily-use acceptance still requires installing these exact public bytes in Codex and Claude Code.
The owner-machine migration and atomic database switch passed on 2026-08-09. The old Pulse and Claude Mem sources, migration copies, and recovery files are preserved in a verified encrypted external archive. Earlier fresh-session checks showed recall, one-call writing, project boundaries, silence on an unrelated question, and fail-open host work in Codex and Claude Code. A later live check on 2026-08-16 exposed the missing reliability gate: the installed epoch 34 daemon reported BGE-M3 ready while ordinary queries timed out. A five-second semantic probe loaded the intact local model and the Hermes query then returned relevant candidates in 496 ms. Daily-use acceptance therefore remains incomplete until the signed 0.8.2 bytes are activated and pass the same cold and warm query checks in both hosts on the owner machine.
The first frozen large-vault baseline on 2026-08-12 did not pass its combined
practical bar. On a copy containing 76,795 real events, published 0.8.0 found
the expected Personal memory in 34 of 40 cases, but stayed silent in only 7 of
10 unrelated controls and returned two internal query errors. Warm p95 was
547 ms and the largest context was about 184 tokens. This makes noise and query
reliability the next defects to fix; it does not invalidate the narrower live
dogfood evidence above. Method and aggregate.
The current product-path benchmark isolates the installed Pulse runtime rather than a Python approximation. The 0.8.1 fix removed weak direct-capsule noise: on the compact product set it recalled 12 of 15 useful memories and stayed silent on all 5 unrelated controls, versus 13 of 15 and 2 of 5 for published 0.8.0. Full raw-history LoCoMo evidence retrieval reached 561 of 1,535 eligible questions. A bounded LongMemEval-S run found an official answer turn for all 30 sampled questions, but that is an oracle retrieval ceiling, not official answer accuracy. The existing emotional benchmark mostly tests hidden user state, biometrics, and graph traversal that Pulse never receives; only 5 of its 35 questions currently test the Personal product. Current benchmark report.
The 0.8.2 preview candidate was also run end to end on all 1,535 eligible LoCoMo questions with the same GPT-5.4 low extraction, answering, and judging model used for equal-model Mem0. It scored 1,230 correct answers (80.13%) versus Mem0's 1,231 (80.20%), up from the previous Pulse baseline of 959 (62.48%). Median context was 120 estimated tokens, maximum context 198, and warm p95 retrieval 70.8 ms. This is strong development evidence, not a production readiness claim. Full candidate result.
A separate remote Claude Chat experiment passed on 2026-08-11 against an
isolated hosted store containing 21 memories. Automatic recall required a
short account-level Instructions for Claude rule and the Pulse connector in
Always available mode; MCP server instructions alone did not trigger the
tool. Each ordinary message therefore makes one visible pulse_recall call.
This proves the remote connector flow only: the hosted store is not synchronized
with the local Personal 0.8 vault and is not part of the published install.
Version 0.7.2 remains the stable release for Macs with Apple Silicon. Version
0.8.2 is published under the npm preview tag for the same public target.
Its exact archive must pass installation and a real semantic query on a clean
Apple Silicon GitHub runner before publication. Intel Mac, Windows, and Linux
are not public support claims yet; fixture and packaging checks on those
targets do not replace live product acceptance.
Ask your AI agent to inspect this repository and explain the changes before it installs anything. The current published Personal installation is:
npx -y @zbs-gg/pulse@0.7.2 init codex
pulse doctor
pulse homeTo try the 0.8 preview explicitly:
npx -y @zbs-gg/pulse@preview init codexThe installer finds Codex, Claude Code, and Cursor, shows every file it will change, and offers to connect all detected programs. To inspect the same plan without changing anything:
pulse init codex --dry-run
pulse init claude-code --dry-run
pulse init cursor --dry-run
# Install into every detected program after reviewing the displayed changes:
pulse init codex --yes
# Or limit installation to one program:
pulse init codex --only codexThe installer keeps one local Personal store and connects every supported AI
program it finds. It does not need Docker or a cloud account. pulse home
opens Memory Home for inspecting, correcting, and deleting local memories.
In the 0.8 preview, Pulse exposes one memory tool, pulse_memory.
The AI program may call it during an ordinary working turn when a durable
decision, preference, open question, project state, correction, or emotional
moment appears. One call accepts at most three short items and does not create
a separate finalizing turn. The older pulse_remember and pulse_graph_delta
names remain compatibility aliases but are not advertised to the model.
Raw conversation capture and old-chat import are off by default. Secret-like, transcript-like, and path-like payloads are rejected. No backend model call is enabled as a hidden default.
Memory stays in the local Pulse data directory. It is not committed to Git and
is not sent to a Pulse server. Disconnecting an AI program preserves memory;
pulse wipe --confirm "wipe pulse memory" is the separate destructive action.
An emotional mark records a short description of the moment, one or more emotions, their strength, and whether the emotion or its cause came from you or was inferred by Pulse. Its influence on the current answer halves every 24 hours and stops after seven days; the historical event remains until you edit or delete it in Memory Home. If a strong emotion has no known cause, Pulse may ask one short question inside the next ordinary answer. It never starts another turn by itself.
To prepare a separate merged Personal database without changing old files:
pulse migrate local-stores --out local-memory-preview.json --open
pulse migrate local-status --jsonOnly actual contradictions require a choice in the local review page. The final switch requires the exact confirmation shown by Pulse and keeps every source database untouched.
Pulse memory is optional. If its daemon or activation is broken, the AI program must still leave the terminal, files, Stop/Cancel, goal controls, and normal session completion available. A failed memory attempt must not create an automatic continuation. The repository tests this boundary, but a packaged test is not a substitute for a fresh real session in each AI program.
The 0.8 preview does not preload a session memory package. Each user
question triggers one temporary local relevance search; the question is not
saved. At most four memories and about 600 tokens are offered, and weak matches
produce no memory context. Stable rules remain in AGENTS.md or CLAUDE.md
instead of being duplicated by Pulse.
The same branch adds bounded storage maintenance. pulse storage reports
protected releases and generated files that can be removed. pulse storage clean requires an exact confirmation, keeps the active release plus one
rollback, and removes only unchanged generated artifacts after a launch and
recall check. It never includes the active vault, keys, migration sources, or
unknown files.
Official 0.6.7 and 0.7.1 Personal databases are upgraded in place without
losing their records. A database made by an unpublished shared-memory build is
refused with a clear error and is left byte-for-byte unchanged. Use the local
merge preview above instead of opening an old mixed database directly.
@zbs-gg/pulse is the open local Personal product. Pulse Team is a separate
private pilot and is not included in this repository or npm package. The
existing AGPL license of Personal code has not been changed by that split.
make verify
make release-verifymake verify builds, formats-checks, and tests the Go engine, local MCP server,
and Personal CLI in an isolated temporary data directory. make release-verify adds the packaged Personal release checks. Neither command may
use ~/.pulse.
The local MCP connection remains stdio. It accepts both the final
2026-07-28 protocol and older clients. Stateless transport does not make the
memory temporary: the data remains in the local SQLite database.
Read AGENTS.md before an agent changes installation or global harness configuration. Security and rollback details are in docs/SECURITY_INSTALL_CHECKLIST.md.
Status: Personal 0.8.2 is the npm preview and its publication gate includes a real semantic query. The owner Mac still runs epoch 34 from 0.8.0, which exposed a cold-search failure while reporting itself ready. The one-day Codex and Claude Code acceptance is therefore not complete until exact public 0.8.2 bytes are installed and pass real recall and writing. Raw-history recall and Cursor acceptance remain pending product limits. A separate Claude Chat remote-recall experiment passed with a visible tool call on every message, but is not Personal sync or a supported install. Version 0.8 is not the stable default or production ready.