Rendu depuis le dépôt source en conservant titres, exemples, code, tableaux, liens et images.
Taste your own cooking: the moment you finish making something, check it with the very skills this repo ships before you serve it.
Goal
A reminder buried in docs ("remember to verify") won't reliably fire in a fresh session. sip makes the recursive self-improvement loop a triggered habit: right after any create or change, run our own skills on the result — the clean checks and the true ones — so quality doesn't ride on the author's biased in-session judgment.
Workflow
- Spot the trigger: you just created or changed an artifact or skill and are about to call it done, commit, or hand it off.
- Cold-read it — run
showeron the artifact (fresh-eyes comprehension / handoff check). - Verify it's true — if the artifact asserts a reality-grounded claim, run
factchk; if it defines an eval, metric, or experiment, runmandela. Skip when it has neither. - Check consistency — run
ssotizein audit mode across the repo for anything the change duplicated or contradicted; execute its consolidation plan only after approval. - Detool portability claims — run
detoolonly when the artifact claims portability, tool-neutrality, stack-agnostic durability, or cross-agent reuse. Skip when it is provenance, operational notes, a tool-targeted runbook, or does not claim portability. - Tidy —
re0the changed docs so the result reads as a clean v0, not a patch over a draft. - Apply the findings here, then serve it.
Rules
- Trigger on your OWN output, right after making it — that's when bias is highest and a check is cheapest.
- Use the skills; don't re-implement them —
showerfor clarity,factchk/mandelafor truth,ssotizefor SSOT,detoolfor portability claims,re0for cleanup.siporchestrates and routes findings back to the author session to fix; the skills do the work. - Skip what plainly doesn't apply, or any check whose skill isn't installed — run only what's present.
factchk/mandelafire only when there is a claim or an eval;detoolfires only when the artifact claims portability, tool-neutrality, stack-agnostic durability, or cross-agent reuse; a one-line prose tweak may need only a consistency check. Say what you skipped and why. - Stop at the artifact —
sipnever touches git or makes commits. - Chain only model-invoked skills; a user-invoked skill (marked
disable-model-invocation) is a human's to fire deliberately, sosipmust not call one.
Verification
Before finishing:
- The change was actually run through the relevant skills, not eyeballed.
- Findings were applied (or consciously deferred with a stated reason).
- The artifact actually changed as a result (a diff exists), or every skipped skill has a stated reason.

