SkillOpt

url/SkillOpt

Fork 0

Commit Graph

Author	SHA1	Message	Date
Yifan Yang	dae974a5e3	chore(sleep): English-only across the engine, plugins, and docs Remove every non-ASCII/CJK character for a professional open-source repo: - harvest.py: drop hardcoded Chinese feedback phrases; add an env-based extensibility hook (SKILLOPT_SLEEP_NEG_FEEDBACK / _POS_FEEDBACK) so any locale can be added without baking one in. Verified with a German example. - rollout.py / consolidate.py: English comments. - README.md section heading + anchor, CONTROLLABLE_DREAMING.md, plugin.json, marketplace.json (also fixed stale path skillopt-sleep-plugin -> plugins/claude-code), SKILL.md: English only. - Remove the internal WAKE_UP_SUMMARY.md note (not user-facing, not referenced). Verified: zero CJK chars remain anywhere; 29 tests pass. Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>	2026-06-08 14:31:52 +00:00
Yifan Yang	e2de84d36f	docs(sleep): real Claude<->Codex cross-validation of the new features Three live runs exercise the new code paths on both runtimes: A) Claude Sonnet->Haiku, gate=OFF + rollouts_k=2: brief-writer test 0->1.00, action 'greedy_improved', val & test both reported (3-way split works). B) Codex, gate=ON + rollouts_k=2: brief-writer test 0->1.00 in 2 nights. C) Claude Sonnet->Haiku, thorough-analyst, 3 nights: slow-update fires and distils a durable cross-night meta-rule (general, not task-specific). Confirms gate-off greedy path, 3-way val/test split, multi-rollout, and the gate-independent slow-update all work with real models on Claude AND Codex. Raw logs under docs/sleep/raw/crosscheck_*.txt. Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>	2026-06-08 14:31:51 +00:00
Yifan Yang	9379e494bf	docs(sleep): document the controllable dreaming architecture Captures the four-stage refactor: train(dream)/val(real)/test(real) splits, optional gate, gate-independent slow-update long-term memory, token/time budget, multi-rollout contrastive reflection, multi-objective reward (accuracy/tokens/ latency), and user-preference priors — with a one-command example composing them. Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>	2026-06-08 14:31:51 +00:00

Author

SHA1

Message

Date

Yifan Yang

dae974a5e3

chore(sleep): English-only across the engine, plugins, and docs

Remove every non-ASCII/CJK character for a professional open-source repo:
  - harvest.py: drop hardcoded Chinese feedback phrases; add an env-based
    extensibility hook (SKILLOPT_SLEEP_NEG_FEEDBACK / _POS_FEEDBACK) so any
    locale can be added without baking one in. Verified with a German example.
  - rollout.py / consolidate.py: English comments.
  - README.md section heading + anchor, CONTROLLABLE_DREAMING.md, plugin.json,
    marketplace.json (also fixed stale path skillopt-sleep-plugin ->
    plugins/claude-code), SKILL.md: English only.
  - Remove the internal WAKE_UP_SUMMARY.md note (not user-facing, not referenced).

Verified: zero CJK chars remain anywhere; 29 tests pass.

Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>

2026-06-08 14:31:52 +00:00

Yifan Yang

e2de84d36f

docs(sleep): real Claude<->Codex cross-validation of the new features

Three live runs exercise the new code paths on both runtimes:
  A) Claude Sonnet->Haiku, gate=OFF + rollouts_k=2: brief-writer test 0->1.00,
     action 'greedy_improved', val & test both reported (3-way split works).
  B) Codex, gate=ON + rollouts_k=2: brief-writer test 0->1.00 in 2 nights.
  C) Claude Sonnet->Haiku, thorough-analyst, 3 nights: slow-update fires and
     distils a durable cross-night meta-rule (general, not task-specific).

Confirms gate-off greedy path, 3-way val/test split, multi-rollout, and the
gate-independent slow-update all work with real models on Claude AND Codex.
Raw logs under docs/sleep/raw/crosscheck_*.txt.

Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>

2026-06-08 14:31:51 +00:00

Yifan Yang

9379e494bf

docs(sleep): document the controllable dreaming architecture

Captures the four-stage refactor: train(dream)/val(real)/test(real) splits,
optional gate, gate-independent slow-update long-term memory, token/time budget,
multi-rollout contrastive reflection, multi-objective reward (accuracy/tokens/
latency), and user-preference priors — with a one-command example composing them.

Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>

2026-06-08 14:31:51 +00:00

3 Commits