tim/parsecapere-prompts
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
13 Commits 1 Branch 0 Tags
d68fdacf2d877f83bef9e9d504149f9f6a127250
Commit Graph

3 Commits

Author SHA1 Message Date
tim d68fdacf2d M3.12: personas/kreuzwort-builder.md — getrimmt auf Domain-Wissen 
Status 0.1.0 (monolithic) → 0.2.0 (trimmed).

Was raus ist:
- Sub-Agent-Intro → base/identity (Stream-Box-Sichtbarkeit)
- "KEIN record_*, KEINE Brick-Manipulation" → workflows/sub-agent-discipline
- "KEINE anderen Tools — du hast NUR die 5 oben" → workflows/tool-scope-lock
- Self-Correction-Loop-Erklaerung → workflows/self-correction
- atomic_finalize-Layer-Pattern → tools-context/atomic-finalize
- "Antworte mit kurzer Success-Meldung" → base/style

Was drin bleibt (Kreuzwort-Domain-Unique):
- Grid-Regeln (max 10x10, 5-8 Woerter, 3-10 Buchstaben, AE/OE/UE/SS,
  Connected-Component, keine Parallel-Beruehrung)
- 6-Schritt-Workflow-Sequenz mit konkreten Tool-Argumenten
- Nummerierungs-Regel fuer Wort-Anfangs-Cells (row-major)
- Final-Message-Format

Frontmatter erweitert um required_prompt_modules + trigger_keywords +
hop_budget — Vorgriff auf M4-Skill-Manifest.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-06-02 15:35:41 +02:00
tim 06859296db M3.11: personas/main-voice.md — getrimmt auf Main-Voice-Unique 
Status 0.1.0 (monolithic) → 0.2.0 (trimmed).
Reduktion: 217 → 92 Zeilen (-58%).

Was raus ist:
- UR-MISSION-Block → base/identity.md
- PLANUNG VOR AKTION → workflows/plan-before-action.md
- ANTWORT-STIL universeller Teil → base/style.md
- Anti-Halluzination universell → base/constraints.md
- BRICK-WORKFLOWS Detail-Patterns → tools-context/brick-creation.md

Was drin bleibt (Main-Voice-Unique):
- USER-SPRACHE → TOOL-SEQUENZ Legende
- add_source_item-Anti-Halluzinations-Regel
- SCOPE & ESCALATION mit wishlist_feature
- Main-Voice-Stil (Plan-Satz, kein Funktionsname)
- SYSTEM_PROMPT_INTRO als Spezialfall-Section am Ende

Frontmatter zeigt jetzt required_prompt_modules + required_tools_high_level
— Vorgriff auf M4-Skill-Manifest.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-06-02 15:35:00 +02:00
tim f4f147e519 M2: 1:1 extraction of monolithic prompts from llm-gateway 
Pulled the System-Prompts out of the three .js files in /root/llm-gateway/
on the Stream-Box-VPS, dropped them 1:1 as Markdown personas — no
modularisation yet (that is M3).

Personas:
- personas/main-voice.md
  ← llm/gemini.js lines 64-172
  (SYSTEM_PROMPT_VOICE + SYSTEM_PROMPT_INTRO)
- personas/kreuzwort-builder.md
  ← orchestrator/kreuzwort-orchestrator.js lines 47-100
- personas/lueckentext-builder.md
  ← orchestrator/lueckentext-orchestrator.js lines 35-68

Plus README.md explaining the target topology
(base/ + workflows/ + tools-context/ + personas/) and a skill-index.md
stub that M4 will fill.

Empty folders kept with .gitkeep so Git tracks the structure even
before M3 starts populating them.

Reference notes:
  arch-flow/brainstorm/Sprache/
    2026-06-02 — Master-Synthese — Prompt-Migration-Reihenfolge.md
    2026-06-02 — Prompt-Modularisierung — Architektur und Composition.md

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-06-02 15:11:42 +02:00