memory
The memory skill manages OpenSquilla's persistent storage system across multiple Markdown files: USER.md for stable profile data, MEMORY.md for curated long-term facts and preferences, and memory/**/*.md for daily or session notes. Use this skill when users request to store, retrieve, update, or delete information they want retained across conversations, employing available tools like memory_search and memory_save to interact with these source files while respecting boundaries around credentials and temporary deliverables.
git clone --depth 1 https://github.com/opensquilla/opensquilla /tmp/memory && cp -r /tmp/memory/src/opensquilla/skills/bundled/memory ~/.claude/skills/memorySKILL.md
# OpenSquilla Memory Use only tools that are visible in the current tool list. This skill explains OpenSquilla's memory source files; it does not make hidden tools available. ## Source Files - `USER.md`: stable user profile fields such as name, preferred address, pronouns, timezone, and durable profile notes. Edit it with visible filesystem tools, not `memory_save`. - `MEMORY.md`: curated long-term non-profile facts, preferences, decisions, and constraints. - `memory/YYYY-MM-DD.md` and `memory/**/*.md`: daily, session, or named memory notes. - `turns/**/*.md`: private auto-captured turn state. These files are for audit/future processing and are not indexed or returned by ordinary `memory_search`. The Markdown files are the source of truth. The memory index/database is derived from curated `MEMORY.md` and `memory/**/*.md` files only. ## Recall - Use injected `USER.md` first for current user identity/profile questions. - Use `memory_search` for historical or non-profile recall that is not already in injected context. - Use `memory_get` after search when exact lines or more context are needed. ## Remember Or Update - If the user specifies a memory path, use that exact path if it is a valid memory source file. - For profile facts, edit `USER.md` with visible filesystem tools. - For daily or session notes, write to `memory/YYYY-MM-DD.md` or another appropriate `memory/**/*.md` source. - For curated long-term facts in `MEMORY.md`, read the current file first and write the full updated content. If `memory_save` is available, use `mode='replace'` for `MEMORY.md`; do not append to it. - If `memory_save` is available, use it only for `MEMORY.md` or `memory/**/*.md`, never for `USER.md`. - If `memory_save` is not available but filesystem tools are visible, edit or create the same source files directly. ## Forget Or Correct - Search first, then read the relevant file/lines before removing anything. - If `memory_delete` is available, use it only when the user wants to delete a whole memory source file. - To remove or correct one fact, edit the source file directly when filesystem tools are visible. - If no write or delete tool is available, report the exact path and lines that should be changed instead of claiming the memory was updated. ## Boundaries - Do not store ordinary deliverables such as reports, JSON outputs, or result files in memory source files. - Do not save secrets, tokens, private keys, or full credential contents. - Only confirm memory was updated after the write or delete succeeds.
Submit audio or video for multilingual dubbing, poll status, and download dubbed audio. Use when the user asks for dubbing, 多语言配音, 视频翻译配音, 译制片, or wants a source clip dubbed into another language.
Generate a structured short-video shooting script from a topic. Emits a strict, machine-parseable shot list (3 shots by default) with image prompt + video prompt + voiceover + on-screen text per shot. Trigger when the user asks for a video script, 分镜, 短视频文案, AI视频, 短剧脚本, or wants visual prompts ready for image/video generation.
Use when the user asks to schedule recurring tasks, one-off reminders, timers, or cron-style jobs through the OpenSquilla cron tool.
Multi-round research with explicit methodology, evidence tracking, and citation-tagged synthesis. Trigger on 'deep dive', 'research report', 'literature review', 'investigate X across sources', 'multi-round investigation'. Distinct from the `summarize` skill, which is a single-pass condensation; this skill maintains a state file across iterations, tracks coverage, and produces a long-form report with per-claim citations. Three execution stages: plan (scope into sub-questions), iterate (record evidence per round), compile (synthesize report). The skill itself does not fetch the web — it tells the host agent which fetches to perform via OpenSquilla's existing web tools, and records what comes back.
Read, edit, or create Microsoft Word `.docx` files. Trigger this skill whenever the user mentions a Word document, .docx file, contract, report, brief, memo, or asks to extract text, modify an existing doc, generate one from a brief, or audit tracked changes. Three execution paths: text-and-structure extraction, in-place edit-by-run (preserves styles), and create-from-scratch with python-docx. Falls back to OOXML unzip-and-patch for layout work python-docx cannot reach.
Capture the current git diff (staged, working-tree, or staged file list) as text. Direct shell call for workflows that need repository diffs without an LLM agent loop.
GitHub operations via `gh` CLI: issues, PRs, CI runs, code review, API queries. Use when: (1) checking PR status or CI, (2) creating/commenting on issues, (3) listing/filtering PRs or issues, (4) viewing run logs. NOT for: complex web UI interactions requiring manual browser flows (use browser tooling when available), bulk operations across many repos (script with gh api), or when gh auth is not configured.
Query the per-turn DecisionEntry log for skill co-occurrence patterns, meta-skill usage stats, and the router fixture corpus. Returns a JSON summary suitable for downstream LLM consumption. Used by meta-skill-creator's harvest step but also useful standalone for 'which skills did I use most this week?'