improve
/improve
Stosuje zebrane sygnały samodoskonalenia: czyta refleksje, wprowadza sprawdzone zmiany i aktualizuje lekcje w pliku pamięci.
- Autor
- slogsdon ↗
- Repozytorium
- loop-and-gate-foundation ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 30.09.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 1
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Brak sygnałów. Repozytorium zmienione 30.09.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Stosuje zebrane sygnały samodoskonalenia: czyta refleksje, wprowadza sprawdzone zmiany i aktualizuje lekcje w pliku pamięci.
Wyzwalacze z opisu
Oryginalny opis (EN)
Apply accumulated self-improvement signals. Use when the user asks to run an improvement pass or "apply reflections" — reads Reflections/, applies changes that have earned it, updates standing lessons in MEMORY.md.
Jak działa
- Steps
- Rules
Podgląd SKILL.md
---
name: improve
description: Apply accumulated self-improvement signals. Use when the user asks to run an improvement pass or "apply reflections" — reads Reflections/, applies changes that have earned it, updates standing lessons in MEMORY.md.
---
# Improve
One job: turn repeated reflection signals into actual changes — skill edits,
MEMORY.md standing lessons, config tweaks. This is where the system compounds:
the model never changes, but the environment it runs in gets sharper.
## Steps
1. Read all notes in `vault/Reflections/` (newest first). Collect every
`Proposed change` block with `status: proposed`.
2. For each proposal, decide:
- **Apply** if the signal repeated (same lesson in 2+ reflections, or
marked REPEAT SIGNAL) AND the target is concrete AND it survives step 2a.
- **Reject** if it appeared once and never again (write `status: rejected`
with one line why), or if it's vague ("be more careful" is not a change).
- **Hold** (leave proposed) if it appeared once and is recent — it may
repeat.
2a. **Verify every proposal that reached Apply.** Use the `verify` skill: one
fresh-context subagent per proposal, given the claim and the evidence paths
(daily notes, vault git diffs, the actual files) and never the Reflections
notes. A `drop` verdict sends the proposal back to Hold with the verifier's
one-line why appended — repeats are not evidence on their own, because the
hook shows each session the last reflection before it writes its own.
3. Apply accepted changes:
- `skills/*/SKILL.md` → edit the skill. Add the rule where it belongs in
the existing structure. Don't append a "lessons" dump at the bottom.
Keep each skill under ~80 lines — if a new rule won't fit, an old rule
must be dropped or merged. Never change a skill's one job.
- `CLAUDE.md` → edit the standing behavior rules. Never touch the
"Hard rules" section.
- `vault/MEMORY.md` → add a one-line entry under `## Standing lessons`
(max 10; if full, demote the least-relevant lesson back to its
Reflections note).
- `config.yaml` → change the value, keep the comment accurate.
4. Mark each applied proposal `status: applied` in its Reflections note.
5. Commit — memory and code live in different repos now:
Stage the files you edited by path in both — never `add -A`, which commits
whatever else is in the tree under your message.
- Skill/`CLAUDE.md`/`config.yaml` edits (this repo):
`git add <the files you edited> && git commit -m "feat: apply self-improvement — <summary>"`.
- The `MEMORY.md` standing-lessons edit lands in the vault (external; see
CLAUDE.md → "Resolving the vault path"). Commit it in the vault's own git
if it has one:
`git -C <base> add MEMORY.md <any Reflections notes you restatused> && git -C <base> commit -m "chore: memory update — standing lessons"`.
Self-modification without a reviewable diff is forbidden.
6. **Release it — committed is not shipped.** A skill edit lands in a source
repo; the agent loads the installed copy. For a plugin skill: bump
`.claude-plugin/plugin.json` and `package.json`, push, then `claude plugin
marketplace update <marketplace>` and `claude plugin update
<plugin>@<marketplace>` (Claude Code) or `pi update git:github.com/<repo>`
(pi). Verify by grepping the NEW installed copy (the plugin cache, or
`~/.pi/agent/git/…`) for the text you added — the source file you
edited proves nothing about what runs. Skip only if the agent loads the
edited file directly.
7. Report: what you applied, what you rejected, what you held — one line each.
Say where each applied change is installed, not just committed.
## Rules
- Never apply a single-occurrence signal. One bad session is noise; the same
problem twice is a pattern. This gate is what stops self-thrashing.
- Never apply a proposal the count alone approved. Two strikes get it to the
verifier; the verifier decides. Counting mentions of a claim is not the same
as checking it.
- Walk the user through each applied diff — improvement passes run
interactively so the human stays in the review loop.
- MEMORY.md updates come LAST — after all skill/CLAUDE.md/config edits,
immediately before the commit; it is the prompt-cache prefix (see CLAUDE.md).
- An applied proposal that was never installed is a held proposal with extra
steps. Two passes in a row have shipped fixes the running agent never saw.
- Skills stay atomic: if a proposal would give a skill a second job, reject
it and propose a new skill in your report instead.
- Do not edit this skill (improve) to weaken its own gates.
Wersje
| Repozytorium | Wynik | Licencja |
|---|---|---|
| Ta strona · slogsdon/loop-and-gate-foundation | 90 | MIT |
| terencebristol/terencebristol-toolkit | 100 | MIT |
| spencerbeggs/effected | 100 | MIT |
| crbnos/carbon | 100 | AGPL-3.0 |
| zixwei/AI-Fleet-skills | 100 | MIT |
| notque/vexjoy-agent | 100 | MIT |
| motiondivision/motion | 100 | MIT |
| relai-ai/agent-eval-loop | 100 | Apache-2.0 |
| jeremylongshore/tons-of-skills-marketplace | 100 | MIT |
Migawka bazy z 13.11.2026 · wersja paczki 68c8a5025516 (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 95Learning
Łączy sygnały i oceny z sesji w wzorce, proponuje zmiany do przeglądu i stosuje zatwierdzone. - 95learn
Zamyka pętlę ulepszeń zachowania: zbiera sygnały, proponuje zmiany do przeglądu i stosuje zatwierdzone poprawki. - 100self-improvement-apply
Stosuje zaakceptowane ustalenia przeglądu samodoskonalenia do skilli jako pojedynczo zatwierdzane edycje i commity. - 100self-improve
Uruchamia sesję samodoskonalenia ChaosEngine: lekcje o środowisku pracy trafiają do zgłoszeń GitHub, a produktowe do kolejki. - 100self-improve
Uruchamia sesję samodoskonalenia ChaosEngine: lekcje o środowisku pracy trafiają do zgłoszeń GitHub, a produktowe do kolejki. - 100apply
Wprowadza zaakceptowane ustalenia przeglądu samodoskonalenia do skilli jako pojedynczo zatwierdzane zmiany i commity.