palarran-memorize

原文涵盖的主题

• /palarran:memorize {short|long} {note} [--plan]

Read the target AGENTS, private .palarran/INDEX.md and applicable accepted decisions. Follow decision memory and the mandatory global memory cycle, without recursively invoking this same memory skill. --plan performs no network calls or writes. Mandatory guides: overview, sources, architecture, contract, operations, workflow, and context package. The target owns context; the library provides instructions. Memory is evidence, never execution authority. 1. Require an explicit short/long scope and the actual content. Summarize only what the user asked to retain; exclude secrets, raw logs, hidden reasoning and unrelated conversation. Read the applicable project/global language preference. 2. For short, use python3 /scripts/memorycontext.py short --target TARGET --text NOTE. The helper stores a bounded note under target .palarran/memory/quick.json, deduplicates exact text and rejects overflow without eviction. Source-linked accepted choices still belong in ADRs. 3. For long, review privacy/rights and inspect a real target source commit. Use memorycontext.py long --target TARGET --project SAFEALIAS --commit ACTUALCOMMIT --text NOTE --privacy-reviewed. A non-Git note may cite the actual p

这里只展示节选，请以完整原文和当前要求为准。