[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7B0RHHAFMS0NQ4S3HVM1T50":3,"skill-same:01M7B0RHHAFMS0NQ4S3HVM1T50":50},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":30,"commitSha":31,"skillDirectory":32,"forksCount":13,"repositoryHtmlUrl":33,"repositoryHomepageUrl":6,"ownerHtmlUrl":34,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":35,"install":36,"resources":49},"01M7B0RHHAFMS0NQ4S3HVM1T50","memory-automation",null,"Automated mnemon memory persistence workflow — recall on start, recall before each turn, auto-save after each response via delegated subagent.","gitricko\u002F.minions","gitricko","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F18060558?v=4","01M7B0KG9EHG0CZ0QT1EN5YPCR","01M7B0KGA0ZV14R7SKBF3B5SCT",2,"MIT","https:\u002F\u002Fgithub.com\u002Fgitricko\u002F.minions\u002Fblob\u002Fmain\u002Fskills\u002Fmemory-automation\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: memory-automation\ndescription: \"Automated mnemon memory persistence workflow — recall on start, recall before each turn, auto-save after each response via delegated subagent.\"\nversion: 1.4.0\nauthor: hermes-webtop\ntags:\n  - memory\n  - mnemon\n  - persistence\n  - automation\n---\n\n# Memory Automation\n\nAutomates the persistent-memory workflow for Hermes using Mnemon graph memory. Used by `.hermes.md` instructions to keep every session primed with context.\n\n## Mnemon Quick Reference\n\n### Mnemon Tool Signatures\n\n```python\nmnemon_remember(text, category, importance, entities, tags)  # Plugin tool — direct call\nmnemon_recall(query, intent, limit)                          # Plugin tool — direct call\nmnemon_forget(insight_id)                                    # Plugin tool — direct call\n```\n\nThese are real Hermes plugin tools, not pseudocode. See [Save Patterns](#save-patterns) for when to use them directly vs via delegated subagent.\n\n### Categories (for `category` parameter)\n\n| Category | When to use |\n|----------|-------------|\n| `fact` | Objective information about the user, project, or environment |\n| `preference` | User's likes, dislikes, stylistic choices, workflow preferences |\n| `decision` | Architectural decisions, tool choices, design decisions |\n| `insight` | Non-obvious findings, troubleshooting steps, workarounds |\n| `context` | Session-level context that's useful across sessions |\n| `general` | Anything that doesn't fit above |\n\n### Importance Levels\n\n| Level | When to use |\n|-------|-------------|\n| 5 | Critical — user identity, core preferences, security constraints |\n| 4 | Important — project goals, recurring preferences, key constraints |\n| 3 | Normal — useful context, typical insights |\n| 2 | Minor — nice-to-know details |\n| 1 | Trivial — barely worth saving (use rarely) |\n\n## Recall Patterns\n\n### Full Context Load (Session Start)\n\n```python\n# Called by the SessionStart hook\nmnemon_recall(query=\"\", intent=\"GENERAL\", limit=20)\n```\n\nThis loads the 20 most relevant past insights into context. No specific query needed — just a broad sweep of everything mnemon has stored.\n\n### Topic-Specific Recall (Before Responding)\n\n```python\n# If the hook returns nothing but the topic references past work:\nmnemon_recall(query=\"\u003Ctopic keywords>\", intent=\"GENERAL\", limit=10)\n```\n\nUse the user's message topic as the query. Include key terms, project names, and any domain-specific vocabulary.\n\n### Intent-Targeted Recall\n\n```python\n# When you need to understand WHY a decision was made:\nmnemon_recall(query=\"\u003Ctopic>\", intent=\"WHY\", limit=5)\n\n# When you need to know WHEN something happened:\nmnemon_recall(query=\"\u003Ctopic>\", intent=\"WHEN\", limit=5)\n\n# When you need info about a specific entity:\nmnemon_recall(query=\"\u003Centity name>\", intent=\"ENTITY\", limit=5)\n```\n\n## Save Patterns\n\n### Direct Tool Call (Preferred for Simple Saves)\n\n`mnemon_remember` is a real Hermes plugin tool exposed by the `gitricko\u002Fhermes-plugin-mnemon` plugin. Call it directly from your toolset — no subagent, no CLI invocation needed:\n\n```python\nmnemon_remember(\n    text=\"User prefers concise responses and bullet points over prose.\",\n    category=\"preference\",\n    importance=4,\n    entities=[\"gitricko\"],\n    tags=[\"communication-style\",\"user-preference\"]\n)\n```\n\n**Use direct calls when:**\n- Saving a single fact or preference inline during a response\n- The save is straightforward — no extraction, no multi-item batch\n- You're already in a turn; just batch the tool call with whatever else you're doing\n\nThe tool wraps the `mnemon` CLI. No MCP server, no subagent.\n\n### Delegated Save (Bulk or After Complex Responses)\n\nFor bulk saves (3+ items) or when context is tight, delegate to a subagent with `toolsets=[\"terminal\"]`. The subagent uses the `mnemon remember` CLI since plugin tools may not propagate to child sessions:\n\n```python\ndelegate_task(\n    goal=\"Save insights to mnemon memory\",\n    context=\"\"\"Save the following to mnemon memory:\n\n1. text=\"User prefers bullet-point summaries over paragraphs.\",\n   category=\"preference\", importance=4,\n   entities=[\"gitricko\"], tags=[\"communication-style\"]\n\n2. text=\"The project uses FastAPI with PostgreSQL.\",\n   category=\"fact\", importance=3,\n   entities=[\"project-name\"], tags=[\"tech-stack\"]\n\nFor each item run: mnemon remember \"\u003Ctext>\" \\\\\n  --cat \u003Ccategory> --imp \u003CN> \\\\\n  --entities e1,e2 \\\\\n  --tags \"t1,t2\"\nBinary: \u002Fusr\u002Flocal\u002Fbin\u002Fmnemon.\"\"\",\n    toolsets=[\"terminal\"]\n)\n```\n\n**Use delegated saves when:**\n- Saving 3+ items at once\n- The save requires extraction or reasoning (e.g. \"pick out all preferences from this dialog\")\n- You're low on context tokens and want the work offloaded\n\n### Decision Guide: Direct vs Delegated\n\n| Condition | Method | Reasoning |\n|-----------|--------|-----------|\n| 1–2 items, simple | Direct `mnemon_remember()` tool call | Zero overhead — inlined with response |\n| 3+ items | Delegate to subagent with CLI | Worth the ~20K token spawn cost for bulk |\n| Save requires extraction\u002Fanalysis | Delegate to subagent | Extraction logic is better offloaded |\n| Already doing other tool calls | Batch `mnemon_remember()` with them | Same turn, no extra cost |\n| `.hermes.md` says \"delegate subagent\" (blanket) | Follow the 1–2 \u002F 3+ split above instead | Blanket delegation burns ~20K tokens per trivial save (e.g. \"tmr\" → timing detail). Proactively update `.hermes.md` to the two-tier system. If user insists on delegation-only, follow their explicit instruction — but default is cost-efficient direct calls. |\n| User explicitly asks for delegation | Always delegate, even for simple saves | User preference overrides efficiency — be transparent about the token cost so they can make an informed choice. |\n\n### Batch Operations with memory() Fallback\n\nFor structured preference data that needs the memory() tool (target='user' or 'memory'):\n\n```python\nmemory(\n    target=\"user\",\n    operations=[\n        {\"action\": \"add\", \"content\": \"User prefers dark mode terminals.\"},\n        {\"action\": \"replace\", \"old_text\": \"old stale fact\", \"content\": \"updated fact\"}\n    ]\n)\n```\n\n## When to Auto-Save (After Every Response)\n\nAfter every user-facing response, scan the exchange for anything on this list:\n\n| Signal | What to save |\n|--------|-------------|\n| User stated a preference | Save as `preference` |\n| User made a decision | Save as `decision` |\n| User corrected you | Save as `preference` or `fact` |\n| You discovered a workaround | Save as `insight` |\n| User revealed personal info | Save as `fact`, importance=5 |\n| User said \"remember this\" | Save whatever follows |\n| Project architecture detail | Save as `fact` |\n| Tool configuration quirk | Save as `insight` |\n\n**When in doubt, save it.** Extra memories are cheap; missing ones are expensive.\n\n## What NOT to Save\n\n- Code the repo already tracks (git history)\n- Public API docs or well-known facts\n- Transient state (\"how are you?\", current time)\n- Things already in `.hermes.md`, `AGENTS.md`, or other config files\n\n## Example: Full Turn Cycle\n\n```\nSession Start → mnemon_recall(\"\", limit=20) → preloaded context\n\nUser: \"I prefer using ruff over black for formatting.\"\n  → Hook: auto mnemon_recall(\"ruff black formatting\", limit=10)\n  → Respond\n  → Direct: mnemon_remember(\n       text=\"User prefers ruff over black for Python formatting.\",\n       category=\"preference\", importance=4,\n       entities=[\"gitricko\"], tags=[\"formatting\",\"python\"]\n     )\n\nUser: \"Remember the project uses Python 3.13\"\n  → Respond\n  → Direct: mnemon_remember(\n       text=\"Project uses Python 3.13.\",\n       category=\"fact\", importance=4,\n       entities=[\"gitricko\"], tags=[\"python\",\"project-setup\"]\n     )\n\nUser: \"Also save these three things: ...\"\n  → Respond\n  → Delegate: bulk save via subagent (3+ items)\n```\n\n## Pitfalls\n\n### `mnemon_remember` IS a Real Tool (Plugin-Exposed)\nThe function `mnemon_remember(...)` is a real Hermes tool, exposed by the `gitricko\u002Fhermes-plugin-mnemon` plugin. Call it directly from your toolset — no CLI needed.\n\n`mnemon_recall(...)` and `mnemon_forget(...)` may also be available depending on what the plugin exposes. Check your tool list at session start.\n\nSubagents may NOT inherit plugin-defined tools. If delegating a save, pass `mnemon remember` CLI commands in context and set `toolsets=[\"terminal\"]` so the subagent runs via shell.\n\n### Real CLI Flags vs Tool Parameters (Subagent Use Only)\n\nWhen delegating to a subagent, it runs the `mnemon` CLI — not the plugin tool. The CLI uses different flags than the tool parameters:\n\n| Tool parameter | CLI flag | Notes |\n|---|---|---|\n| `text` | positional arg | First argument, not a flag |\n| `category` | `--cat` | e.g. `--cat preference` |\n| `importance` | `--imp` | e.g. `--imp 4` |\n| `entities` | `--entities` | Comma-separated: `entity1,entity2` |\n| `tags` | `--tags` | Comma-separated: `tag1,tag2` |\n\n### Sub-Agent Needs `terminal` Toolset\nDelegated saves require `toolsets=[\"terminal\"]` — the sub-agent runs `mnemon` via the shell since plugin tools don't propagate to child sessions. Omitting this or using a wrong toolset leaves the sub-agent with no way to run the CLI.\n\n### Load This Skill Proactively\nThis skill must be loaded at the start of every session (via `skill_view(name='memory-automation')`). Do not rely on `.hermes.md` reminders to trigger it — if the skill isn't loaded, you won't follow the workflow. Add it to the start-of-session routine.\n\n### Don't Confuse `mnemon_remember` with `memory()`\n`mnemon_remember()` (plugin tool) → stores to **Mnemon graph DB** — durable, queriable, multi-session.\n\n`memory()` (built-in Hermes tool) → stores to **agent memory** — injected every turn, bounded at 2.2K chars.\n\nThey target different stores. Use `mnemon_remember()` for Mnemon entries and `memory()` for the Hermes built-in memory. The `[MEMORY]` and `[USER PROFILE]` sections injected at the top of every prompt are rendered from the `memory()` tool's data — they are NOT part of `.hermes.md`. Don't conflate the two tools or their targets.\n\n### `.hermes.md` Says Delegate But It's Wasting Tokens — What To Do\n`.hermes.md` often says \"delegate subagent to mnemon_remember\" as a blanket instruction. The skill's Decision Guide says direct calls are fine for simple saves. Follow the Decision Guide's cost-aware split instead (direct for 1–2 simple items, delegate for 3+\u002Fcomplex). Propose updating `.hermes.md` to the two-tier system. If the user explicitly prefers delegation-only, defer — but be transparent about the ~20K token cost per spawn so it's an informed choice.","# Memory Automation\n\nAutomates the persistent-memory workflow for Hermes using Mnemon graph memory. Used by `.hermes.md` instructions to keep every session primed with context.\n\n## Mnemon Quick Reference\n\n### Mnemon Tool Signatures\n\n```python\nmnemon_remember(text, category, importance, entities, tags)  # Plugin tool — direct call\nmnemon_recall(query, intent, limit)                          # Plugin tool — direct call\nmnemon_forget(insight_id)                                    # Plugin tool — direct call\n```\n\nThese are real Hermes plugin tools, not pseudocode. See [Save Patterns](#save-patterns) for when to use them directly vs via delegated subagent.\n\n### Categories (for `category` parameter)\n\n| Category | When to use |\n|----------|-------------|\n| `fact` | Objective information about the user, project, or environment |\n| `preference` | User's likes, dislikes, stylistic choices, workflow preferences |\n| `decision` | Architectural decisions, tool choices, design decisions |\n| `insight` | Non-obvious findings, troubleshooting steps, workarounds |\n| `context` | Session-level context that's useful across sessions |\n| `general` | Anything that doesn't fit above |\n\n### Importance Levels\n\n| Level | When to use |\n|-------|-------------|\n| 5 | Critical — user identity, core preferences, security constraints |\n| 4 | Important — project goals, recurring preferences, key constraints |\n| 3 | Normal — useful context, typical insights |\n| 2 | Minor — nice-to-know details |\n| 1 | Trivial — barely worth saving (use rarely) |\n\n## Recall Patterns\n\n### Full Context Load (Session Start)\n\n```python\n# Called by the SessionStart hook\nmnemon_recall(query=\"\", intent=\"GENERAL\", limit=20)\n```\n\nThis loads the 20 most relevant past insights into context. No specific query needed — just a broad sweep of everything mnemon has stored.\n\n### Topic-Specific Recall (Before Responding)\n\n```python\n# If the hook returns nothing but the topic references past work:\nmnemon_recall(query=\"\u003Ctopic keywords>\", intent=\"GENERAL\", limit=10)\n```\n\nUse the user's message topic as the query. Include key terms, project names, and any domain-specific vocabulary.\n\n### Intent-Targeted Recall\n\n```python\n# When you need to understand WHY a decision was made:\nmnemon_recall(query=\"\u003Ctopic>\", intent=\"WHY\", limit=5)\n\n# When you need to know WHEN something happened:\nmnemon_recall(query=\"\u003Ctopic>\", intent=\"WHEN\", limit=5)\n\n# When you need info about a specific entity:\nmnemon_recall(query=\"\u003Centity name>\", intent=\"ENTITY\", limit=5)\n```\n\n## Save Patterns\n\n### Direct Tool Call (Preferred for Simple Saves)\n\n`mnemon_remember` is a real Hermes plugin tool exposed by the `gitricko\u002Fhermes-plugin-mnemon` plugin. Call it directly from your toolset — no subagent, no CLI invocation needed:\n\n```python\nmnemon_remember(\n    text=\"User prefers concise responses and bullet points over prose.\",\n    category=\"preference\",\n    importance=4,\n    entities=[\"gitricko\"],\n    tags=[\"communication-style\",\"user-preference\"]\n)\n```\n\n**Use direct calls when:**\n- Saving a single fact or preference inline during a response\n- The save is straightforward — no extraction, no multi-item batch\n- You're already in a turn; just batch the tool call with whatever else you're doing\n\nThe tool wraps the `mnemon` CLI. No MCP server, no subagent.\n\n### Delegated Save (Bulk or After Complex Responses)\n\nFor bulk saves (3+ items) or when context is tight, delegate to a subagent with `toolsets=[\"terminal\"]`. The subagent uses the `mnemon remember` CLI since plugin tools may not propagate to child sessions:\n\n```python\ndelegate_task(\n    goal=\"Save insights to mnemon memory\",\n    context=\"\"\"Save the following to mnemon memory:\n\n1. text=\"User prefers bullet-point summaries over paragraphs.\",\n   category=\"preference\", importance=4,\n   entities=[\"gitricko\"], tags=[\"communication-style\"]\n\n2. text=\"The project uses FastAPI with PostgreSQL.\",\n   category=\"fact\", importance=3,\n   entities=[\"project-name\"], tags=[\"tech-stack\"]\n\nFor each item run: mnemon remember \"\u003Ctext>\" \\\\\n  --cat \u003Ccategory> --imp \u003CN> \\\\\n  --entities e1,e2 \\\\\n  --tags \"t1,t2\"\nBinary: \u002Fusr\u002Flocal\u002Fbin\u002Fmnemon.\"\"\",\n    toolsets=[\"terminal\"]\n)\n```\n\n**Use delegated saves when:**\n- Saving 3+ items at once\n- The save requires extraction or reasoning (e.g. \"pick out all preferences from this dialog\")\n- You're low on context tokens and want the work offloaded\n\n### Decision Guide: Direct vs Delegated\n\n| Condition | Method | Reasoning |\n|-----------|--------|-----------|\n| 1–2 items, simple | Direct `mnemon_remember()` tool call | Zero overhead — inlined with response |\n| 3+ items | Delegate to subagent with CLI | Worth the ~20K token spawn cost for bulk |\n| Save requires extraction\u002Fanalysis | Delegate to subagent | Extraction logic is better offloaded |\n| Already doing other tool calls | Batch `mnemon_remember()` with them | Same turn, no extra cost |\n| `.hermes.md` says \"delegate subagent\" (blanket) | Follow the 1–2 \u002F 3+ split above instead | Blanket delegation burns ~20K tokens per trivial save (e.g. \"tmr\" → timing detail). Proactively update `.hermes.md` to the two-tier system. If user insists on delegation-only, follow their explicit instruction — but default is cost-efficient direct calls. |\n| User explicitly asks for delegation | Always delegate, even for simple saves | User preference overrides efficiency — be transparent about the token cost so they can make an informed choice. |\n\n### Batch Operations with memory() Fallback\n\nFor structured preference data that needs the memory() tool (target='user' or 'memory'):\n\n```python\nmemory(\n    target=\"user\",\n    operations=[\n        {\"action\": \"add\", \"content\": \"User prefers dark mode terminals.\"},\n        {\"action\": \"replace\", \"old_text\": \"old stale fact\", \"content\": \"updated fact\"}\n    ]\n)\n```\n\n## When to Auto-Save (After Every Response)\n\nAfter every user-facing response, scan the exchange for anything on this list:\n\n| Signal | What to save |\n|--------|-------------|\n| User stated a preference | Save as `preference` |\n| User made a decision | Save as `decision` |\n| User corrected you | Save as `preference` or `fact` |\n| You discovered a workaround | Save as `insight` |\n| User revealed personal info | Save as `fact`, importance=5 |\n| User said \"remember this\" | Save whatever follows |\n| Project architecture detail | Save as `fact` |\n| Tool configuration quirk | Save as `insight` |\n\n**When in doubt, save it.** Extra memories are cheap; missing ones are expensive.\n\n## What NOT to Save\n\n- Code the repo already tracks (git history)\n- Public API docs or well-known facts\n- Transient state (\"how are you?\", current time)\n- Things already in `.hermes.md`, `AGENTS.md`, or other config files\n\n## Example: Full Turn Cycle\n\n```\nSession Start → mnemon_recall(\"\", limit=20) → preloaded context\n\nUser: \"I prefer using ruff over black for formatting.\"\n  → Hook: auto mnemon_recall(\"ruff black formatting\", limit=10)\n  → Respond\n  → Direct: mnemon_remember(\n       text=\"User prefers ruff over black for Python formatting.\",\n       category=\"preference\", importance=4,\n       entities=[\"gitricko\"], tags=[\"formatting\",\"python\"]\n     )\n\nUser: \"Remember the project uses Python 3.13\"\n  → Respond\n  → Direct: mnemon_remember(\n       text=\"Project uses Python 3.13.\",\n       category=\"fact\", importance=4,\n       entities=[\"gitricko\"], tags=[\"python\",\"project-setup\"]\n     )\n\nUser: \"Also save these three things: ...\"\n  → Respond\n  → Delegate: bulk save via subagent (3+ items)\n```\n\n## Pitfalls\n\n### `mnemon_remember` IS a Real Tool (Plugin-Exposed)\nThe function `mnemon_remember(...)` is a real Hermes tool, exposed by the `gitricko\u002Fhermes-plugin-mnemon` plugin. Call it directly from your toolset — no CLI needed.\n\n`mnemon_recall(...)` and `mnemon_forget(...)` may also be available depending on what the plugin exposes. Check your tool list at session start.\n\nSubagents may NOT inherit plugin-defined tools. If delegating a save, pass `mnemon remember` CLI commands in context and set `toolsets=[\"terminal\"]` so the subagent runs via shell.\n\n### Real CLI Flags vs Tool Parameters (Subagent Use Only)\n\nWhen delegating to a subagent, it runs the `mnemon` CLI — not the plugin tool. The CLI uses different flags than the tool parameters:\n\n| Tool parameter | CLI flag | Notes |\n|---|---|---|\n| `text` | positional arg | First argument, not a flag |\n| `category` | `--cat` | e.g. `--cat preference` |\n| `importance` | `--imp` | e.g. `--imp 4` |\n| `entities` | `--entities` | Comma-separated: `entity1,entity2` |\n| `tags` | `--tags` | Comma-separated: `tag1,tag2` |\n\n### Sub-Agent Needs `terminal` Toolset\nDelegated saves require `toolsets=[\"terminal\"]` — the sub-agent runs `mnemon` via the shell since plugin tools don't propagate to child sessions. Omitting this or using a wrong toolset leaves the sub-agent with no way to run the CLI.\n\n### Load This Skill Proactively\nThis skill must be loaded at the start of every session (via `skill_view(name='memory-automation')`). Do not rely on `.hermes.md` reminders to trigger it — if the skill isn't loaded, you won't follow the workflow. Add it to the start-of-session routine.\n\n### Don't Confuse `mnemon_remember` with `memory()`\n`mnemon_remember()` (plugin tool) → stores to **Mnemon graph DB** — durable, queriable, multi-session.\n\n`memory()` (built-in Hermes tool) → stores to **agent memory** — injected every turn, bounded at 2.2K chars.\n\nThey target different stores. Use `mnemon_remember()` for Mnemon entries and `memory()` for the Hermes built-in memory. The `[MEMORY]` and `[USER PROFILE]` sections injected at the top of every prompt are rendered from the `memory()` tool's data — they are NOT part of `.hermes.md`. Don't conflate the two tools or their targets.\n\n### `.hermes.md` Says Delegate But It's Wasting Tokens — What To Do\n`.hermes.md` often says \"delegate subagent to mnemon_remember\" as a blanket instruction. The skill's Decision Guide says direct calls are fine for simple saves. Follow the Decision Guide's cost-aware split instead (direct for 1–2 simple items, delegate for 3+\u002Fcomplex). Propose updating `.hermes.md` to the two-tier system. If the user explicitly prefers delegation-only, defer — but be transparent about the ~20K token cost per spawn so it's an informed choice.",{"name":5,"tags":23,"author":28,"version":29,"description":7},[24,25,26,27],"memory","mnemon","persistence","automation","hermes-webtop","1.4.0","main","e95aec76487a22bbbc23b63234724bc7178d4692","skills\u002Fmemory-automation","https:\u002F\u002Fgithub.com\u002Fgitricko\u002F.minions","https:\u002F\u002Fgithub.com\u002Fgitricko","2026-11-13 16:00:00",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":48},[38,41,44],{"tool":39,"command":40},"npx","npx skills add gitricko\u002F.minions --skill memory-automation",{"tool":42,"command":43},"bunx","bunx skills add gitricko\u002F.minions --skill memory-automation",{"tool":45,"command":46},"pnpm","pnpm dlx skills add gitricko\u002F.minions --skill memory-automation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgitricko\u002F.minions\u002Fblob\u002Fmain\u002Fskills\u002Fmemory-automation\u002FSKILL.md",true,[],[51,58,65,72,79,86],{"skillNo":52,"name":53,"nameZh":6,"description":54,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":57,"publishedAt":18},"01M7B0VJ3Q0YMCTJY69FEJ4100","version-single-source","Use when adding a new component dependency or versioning a binary in install\u002Fboot scripts.","https:\u002F\u002Fgithub.com\u002Fgitricko\u002F.minions\u002Fblob\u002Fmain\u002Fskills\u002Fversion-single-source\u002FSKILL.md",[],[],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":18},"01M7B0V888DV8J3JY5W9Z6WCF3","tdd","Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions \"red-green-refactor\", or wants integration tests.","https:\u002F\u002Fgithub.com\u002Fgitricko\u002F.minions\u002Fblob\u002Fmain\u002Fskills\u002Ftdd\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":18},"01M7B0TVFM38JH3PBA2K219Z2X","shell-path-login","Use when a binary or script works in one shell context but not another (interactive vs non-interactive login vs non-login shell).","https:\u002F\u002Fgithub.com\u002Fgitricko\u002F.minions\u002Fblob\u002Fmain\u002Fskills\u002Fshell-path-login\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M7B0TKFCTVQBTSNF9WF1MXJ6","self-update-fix","Fix a failing dependency update PR. Use when a self-update PR's CI fails, or when invoked by scripts\u002Fself-update-fix.sh.","https:\u002F\u002Fgithub.com\u002Fgitricko\u002F.minions\u002Fblob\u002Fmain\u002Fskills\u002Fself-update-fix\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M7B0TB57WEW0E6E9874K3C72","pr-change-log","Use when a multi-commit PR needs its description written\u002Fupdated, or when closing out a long branch with many fix commits.","https:\u002F\u002Fgithub.com\u002Fgitricko\u002F.minions\u002Fblob\u002Fmain\u002Fskills\u002Fpr-change-log\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":18},"01M7B0T2FGSV0ZFRZ46PG4BBY2","persistent-knowledge","Persistent skills\u002Fknowledge in Codespace via symlinks.","https:\u002F\u002Fgithub.com\u002Fgitricko\u002F.minions\u002Fblob\u002Fmain\u002Fskills\u002Fpersistent-knowledge\u002FSKILL.md",[],[]]