59b5ebc2d85bdf53d38b6bcf347c496dd3432be1
4
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 JesseMarkowitzandClaude Opus 5
|
59b5ebc2d8 |
v1.1 WP-C: browser release coverage
Drives in a real browser the reader workflows v1 proved only through the API or the component suite, including an export that leaves the browser as a file. Final run: 91 checks (the 38 existing M11 checks plus 53 new), 0 failed, 0 skipped, on the production build over trusted-LAN HTTPS. - tools/m11_browser.py: scenarios for Retry and takes, Save Point create / restore / Redo, state correction (accepted, and a refused correction with its reason), narration length reaching each turn's prompt, failed generation (an unserved model blocked up front; a listed model that cannot narrate failing in the open) and recovery, and export download from the library and from campaign settings, imported into a fresh application. Rows are tagged M11 / WP-C and counted separately; --only for development. The M11 checks now wait on conditions instead of sleeping. - tools/m11_webdriver.py: Firefox download preferences, a $HOME-only download folder, a download wait that ignores partial, empty, pre-existing and still-growing files, centred real clicks, tabs, and condition waits. - tests/test_v11_c_browser_helpers.py: the download wait, prefs and $HOME guard, without a browser. - frontend: a correction the story refused was presented as "Generation failed" with a Retry offer and a typed-input claim. It is now "That correction was not applied", not retryable, with the reason kept (errors.js, FailureNotice.jsx; 3 regression tests). - DEVELOPMENT.md: the harness command, download profile and $HOME rule, what counts as a finished download, and the no-sleep rule. - docs: V1.1-PLAN, VERSION v4.4, planning README, reports/v1.1/V1.1-WP-C-REPORT.md. Open for the owner: "Correct" on an Important Facts row is always refused (K1), and a partly refused correction is not reachable from the reader UI. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VvegagkhuCZoFPdv4M1egY |
||
|
JesseMarkowitzandClaude Opus 5
|
0c1ba836ba |
v1.1 WP-B.2: independent long-term memory retention
Corrects the memory mechanisms WP-B.1 diagnosed, one at a time, each verified before the next. Accepted by the owner with a documented reference-model limitation. No schema, bundle format, setting default, lineage, authority or protocol-cleanup change. - B2.1 ranking: the retrieval query is the player's input plus a bounded scene context (state scene + end of the newest narration), embedded in one call. final = semantic (0.6 input / 0.4 context) + 0.15 x lexical, where lexical is a rarity-weighted share of the input's words, computed per turn over the candidates with no index. Scores and the query are recorded per used memory; pins and redundancy suppression unchanged. - B2.2 coverage-aware eviction (memorybank.eviction_order): the earliest and newest memories are kept, the smallest coverage hole goes first, least-recently-used breaks ties and remains the fallback. Bounded; pins never evicted; frozen-bank protection kept; reads no text or vectors. - B2.3 bounded memory creation: a block longer than 2,000 tokens is shown to the summariser as head + tail with an omission marker, inside the same budget; shorter blocks unchanged; the marker is never stored. - The memory summariser prompt is unchanged from v1.0.0. A B2.4 prompt experiment was measured on the reference model, showed no reliable improvement for the target failure (0/5 under both prompts, with new "Memory:"-prefix, second-person and length regressions), and was reverted. memorybank.memory_user_prompt is kept as a behaviour-neutral helper. - tools/memory_fidelity.py (diagnostic only): genre-neutral fixtures plus the failed block, a deterministic fidelity checker, and a real-model shipped-vs-experiment measurement. - tools/memory_diagnostic.py: ranking replica uses production scoring; ranking_crowded, ranking_context_dependent and independent_full fixtures; per-turn isolation and provenance. - tests: B.1's two strict xfails are now ordinary passes; ranking, eviction and excerpt tests; summariser acceptance tests kept apart from diagnostic-measurement tests. - DEVELOPMENT.md: the GPU-host kernel/Ollama watch used `-k -u ollama`, which matches nothing; now the OR form. - docs: CONTEXT-AND-MEMORY 15/18/20/21 as shipped, V1.1-PLAN (status and release criteria 12-13), planning README, VERSION v4.3, reports/v1.1/V1.1-WP-B2-REPORT.md. Deterministic independent-memory recovery: PASS (independent_full fails on v1.0.0 at creation and returns recovered_through_memory_independent here). Reference-model independent recovery: FAILED on the precondition-valid attempt, at memory creation: the summariser omitted a player-established fact from a block it received whole. Accepted as a documented v1.1 residual and carried into the release gate. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VvegagkhuCZoFPdv4M1egY |
||
|
JesseMarkowitzandClaude Opus 5
|
beb17ada10 |
v1.1 WP-B.1: diagnose independent long-term memory retention
Diagnostic only; no memory behaviour changes. - tools/memory_diagnostic.py: planted-fact isolation checks, the four-stage diagnosis (created / retained / ranked / injected) with a verdict, a production-ranking replica, deterministic summariser/embedder/narrator stubs and seven scenarios (default, past capacity, pinned, low top_k, long-block early/late, lineage control) - tools/v11_b1_memory.py: CLI for the scenarios and for diagnosing a copy of a finished real campaign - tools/m11_long_run.py: opt-in --independent-fact mode with per-turn isolation tracking and the recovered_through_memory_independent verdict; M04 verdicts unchanged - tests: diagnostic stages, eviction, creation window, ranking, lineage and authority controls; two strict xfails record the diagnosed retention and creation defects for WP-B.2 to flip - planning/reports/v1.1/V1.1-WP-B1-REPORT.md First failing stage: ranking (real model); retention past capacity and creation for early facts in long blocks (deterministic, same on v1.0.0). Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VvegagkhuCZoFPdv4M1egY |
||
|
JesseMarkowitzandClaude Opus 5
|
d63804f22e |
v1.1: harden context window and narrator protocol boundary
WP-A1 and WP-A2, implemented in sequence, plus the corrective work the owner
asked for at review. Reported in
planning/reports/v1.1/V1.1-WP-A1-A2-REPORT.md (corrective addendum §R).
Planning package v4.2.
WP-A1: context-window safety reserve
- The prompt leaves max(256, ceil(5% of the effective window)) tokens free
beside the reply. That is 256 at 4,096 and 820 at 16,384. The value is fixed,
not a setting, and not calibrated per model.
- M6's 64-token margin is gone. Separators and the chat hint are priced
exactly; tokenizer drift is the reserve's job.
- Protected context that cannot fit raises ContextOverflow before the model
is called.
- Streams set stream_options.include_usage. Measured on Ollama 0.33, a stream
sent no usage without it.
- Each sent turn records fits, exceeded, truncation_suspected or unknown.
The status is returned on the done event, logged when bad, and shown in the
context inspector. The turn is always kept.
- Accounting is per-attempt data (attempts.ATTEMPT_KEYS).
- Corrective: a cold model is loaded before its turn is built. When the
window is unverified but the server answered, contextwindow.ensure_window
makes one bounded POST /api/generate naming only the model. It sends no
prompt, generates nothing and writes nothing. It then probes again, and the
turn is built to that answer. If the load fails, or the window is still
unknown, the turn falls back to the old behaviour.
- Real host, 4,096 window:
- v1 cold turn: sent 13,875, the server read 2,050.
- Same turn after the correction: the window was verified, 3,082 sent,
3,097 read, fits, 499 tokens left beside the reply.
- Verified turns elsewhere left 275-2,297 tokens against v1's 23-42.
WP-A2: protocol echo and genre-neutral state prompting
- The vocabulary is shown as the JSON object the model sends, not as
name(field, ...). This costs 121 tokens.
- The example uses character-1, item-1 and location-1.
- The extractor removes shapes anchored to application-owned text:
- a vocabulary call line;
- an echoed length hint;
- the renderer's scene line left last;
- an empty fence opener.
- Corrective R5: the echoed continue hint is recognised by its own sentence
("Output only story text"). A Hard-limit-opened bracket is removed only
directly above an echo already cut from the same reply.
- Replay of all 518 real v1 replies: 9 changed, 0 flagged, and no story prose
removed. That is unchanged by R5.
- Replay of 64 v1.1 replies: 3 changed, 0 flagged. The depth-16 instruction
tail is removed.
- Identity diagnostic after the correction:
- 0 identity signals;
- 0 prompt example identifiers proposed;
- 0/10 stored turns with protocol or instruction shapes.
- 50-turn run: 51 accepted, 0 of 54 stored turns carry protocol.
- SPECS, render.py and validate.py are identical to v1.0.0.
Compatibility: a real v1.0.0 database reads identically on v1.0.0 and v1.1,
field for field, with schema and user_version 94 unchanged. Undo, redo, Save
Point restore, export and import all work on it. There is no schema,
migration or bundle-format change.
Verification: the backend suite passes 1,534 with 17 skipped and 0 failed.
The frontend passes 165/165, and lint and the build are clean. The offline
container and the browser regression were re-run on this tree (see §R.3).
One test was re-calibrated, not weakened: test_history_block_trim's prefix
test had assumed which turn holds the floor at a 2,048 budget.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VvegagkhuCZoFPdv4M1egY
|