[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7ETMZDXVBFZBYPS8DA5KWT9":3,"skill-same:01M7ETMZDXVBFZBYPS8DA5KWT9":78},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":38,"commitSha":39,"skillDirectory":40,"forksCount":41,"repositoryHtmlUrl":42,"repositoryHomepageUrl":6,"ownerHtmlUrl":43,"ownerBlogUrl":6,"ownerDescription":44,"updatedAt":45,"install":46,"resources":59},"01M7ETMZDXVBFZBYPS8DA5KWT9","threat-model",null,"Build a threat model for a target codebase. Three modes: \"interview\" walks an application owner through the four-question framework and produces a threat model from their answers; \"bootstrap\" derives a threat model from the code plus past vulnerabilities (CVEs, git history, pentest reports) when no owner is available; \"bootstrap-then-interview\" chains the two when both owner and codebase are present. All write THREAT_MODEL.md in a shared schema. Use when asked to \"threat model\", \"build a threat model\", \"map the attack surface\", or \"what should we be worried about in this codebase\".","lza6\u002FArmor-piercing-system-prompts","lza6","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F128880206?v=4","01M7EB3K7WPX5T28EJ2NVHE6EG","Chinese-tingfeng","01M7EB3K97E1KQ16QWKJMFNGX8",11,"MIT","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fthreat-model\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: threat-model\ndescription: >-\n  Build a threat model for a target codebase. Three modes: \"interview\" walks an\n  application owner through the four-question framework and produces a threat\n  model from their answers; \"bootstrap\" derives a threat model from the code\n  plus past vulnerabilities (CVEs, git history, pentest reports) when no owner\n  is available; \"bootstrap-then-interview\" chains the two when both owner and\n  codebase are present. All write THREAT_MODEL.md in a shared schema. Use\n  when asked to \"threat\n  model\", \"build a threat model\", \"map the attack surface\", or \"what should we\n  be worried about in this codebase\".\nargument-hint: \"[bootstrap-then-interview|bootstrap|interview] \u003Ctarget-dir> [--vulns \u003Cfile>] [--design-doc \u003Cfile>] [--seed \u003CTHREAT_MODEL.md>] [--fresh]\"\nallowed-tools:\n  - Read\n  - Glob\n  - Bash(python3 .claude\u002Fskills\u002F_lib\u002Fcheckpoint.py:*)\n  - Grep\n  - Write\n  - Bash(git:*)\n  - Bash(gh api:*)\n  - Bash(find:*)\n  - Bash(ls:*)\n  - Bash(cat:*)\n  - AskUserQuestion\n  - Task\n---\n\n# threat-model\n\nA threat model answers **\"what could go wrong with this system, who would do\nit, and what should we do about it?\"** independently of whether any specific\nbug has been found yet. It is the map; vulnerability discovery is the metal\ndetector. A good threat model tells the pipeline where to look and tells triage\nwhich findings matter.\n\n**Litmus test:** If patching one line of code makes an entry disappear, it was\na vulnerability, not a threat. A threat (\"attacker achieves RCE via untrusted\nmedia parsing\") still stands after every known bug is fixed; a vulnerability\n(\"`dr_wav.h:412` doesn't bounds-check `chunk_size`\") does not. This skill\nproduces threats. Vulnerabilities appear only as **evidence** that raises a\nthreat's likelihood score.\n\n**Invocation:** `\u002Fthreat-model [bootstrap-then-interview|bootstrap|interview] \u003Ctarget-dir> [flags]`\n\n---\n\n## Step 0 — Safety preamble (always runs first)\n\nThis skill performs **static analysis only**. It reads source, git history,\nand any vulnerability reports the user supplies, and writes a single output\nfile (`\u003Ctarget-dir>\u002FTHREAT_MODEL.md`). It does not build, execute, fuzz, or\nmodify the target, and does not make network requests against the target's\ninfrastructure.\n\nBefore proceeding, confirm and state in your first response:\n\n1. The target directory exists and is a local checkout you can read.\n2. You will not execute any code from the target directory.\n3. If `--vulns` points at a URL or you are asked to \"fetch CVEs\", you will\n   query only public advisory databases (NVD, GitHub Security Advisories, the\n   project's own issue tracker) and never the target's live deployment.\n\nIf the user asks you to validate a threat by running an exploit, decline and\npoint them at the `vuln-pipeline` (README Step 2) instead.\n\n---\n\n## Step 1 — Route to a mode\n\nParse `$ARGUMENTS`:\n\n| First token | Route to |\n|---|---|\n| `interview` | Read `interview.md` in this directory and follow it. |\n| `bootstrap` | Read `bootstrap.md` in this directory and follow it. |\n| `bootstrap-then-interview` | Bootstrap first, then interview seeded from the draft. See below. |\n| anything else, or empty | Ask the user: **\"Is someone who owns or built this system available to answer questions in this session?\"** Yes and the codebase is checked out → recommend `bootstrap-then-interview`. Yes but no codebase → `interview.md`. No → `bootstrap.md`. |\n\nAll modes write the same artifact (`THREAT_MODEL.md`, schema in `schema.md`)\nso downstream consumers (pipeline `recon`\u002F`judge`, verifier agents) do not need\nto know which mode produced it.\n\n| | `interview` | `bootstrap` |\n|---|---|---|\n| **Needs** | An application owner present in the session | A local checkout; optionally past vulns |\n| **Method** | Four-question framework: conversational walk through *what are we working on → what can go wrong → what are we going to do about it → did we do a good job* | Five stages: parallel research swarm → synthesize sections 1-3 + vuln table → generalize vulns into threat classes → STRIDE gap-fill → emit |\n| **Best for** | New systems, design reviews, systems where the risk lives in business logic the code doesn't show | Inherited systems, third-party code, OSS dependencies, anything with a CVE history |\n| **Provenance tag** | `interview` | `bootstrap` |\n\n**Context durability.** Interview mode is multi-turn; tool results from early\nreads may be evicted before you need them. To stay resilient:\n\n- Do **not** read `interview.md` or `bootstrap.md` in full up front. Read the\n  mode file (or the relevant section of it) **at the point you need it**, one\n  question or stage at a time.\n- If a re-read via the Read tool is refused as \"file unchanged\", the prior\n  result was evicted; reload with `cat \u003Cpath>` via Bash instead.\n\n**Interview backbone** (so you can proceed even if `interview.md` is\nunavailable mid-session):\n\n| Q | Question | Fills schema sections |\n|---|---|---|\n| Q1 | What are we working on? | section 1 context, section 2 assets, section 3 entry points |\n| Q2 | What can go wrong? | section 4 threat rows (id, threat, actor, surface, asset) |\n| Q3 | What are we going to do about it? | section 4 impact\u002Flikelihood\u002Fstatus\u002Fcontrols; section 5 deprioritized; section 8 recommended mitigations |\n| Q4 | Did we do a good job? | validate ranking, coverage check, section 6 open questions |\n\n### `bootstrap-then-interview` mode\n\nWhen the owner is available *and* the codebase is checked out, this is the\nrecommended path: the owner's time goes to refining a code-grounded draft\ninstead of describing the system from scratch.\n\n1. Tell the owner: \"I'll read the code first and come back with a draft\n   (about 5-10 min), then we'll walk it together. Want that, or would you\n   rather start cold?\" Only proceed if they opt in; otherwise fall back to\n   `interview.md`.\n2. Read `bootstrap.md` and follow it end-to-end. Write\n   `\u003Ctarget-dir>\u002FTHREAT_MODEL.md`.\n3. Immediately continue into interview mode: read `interview.md` and follow\n   it with `--seed \u003Ctarget-dir>\u002FTHREAT_MODEL.md` in effect. The section 6 open\n   questions from bootstrap become your Q1-Q4 prompts; the owner confirms,\n   corrects, and adds rather than starting from nothing.\n4. Overwrite `\u003Ctarget-dir>\u002FTHREAT_MODEL.md` with the refined model. Set\n   provenance `mode: bootstrap-then-interview`.\n\nThe same flow is available manually: run `bootstrap` first, then\n`interview --seed \u003CTHREAT_MODEL.md>` in a later session.\n\n---\n\n## Step 2 — Shared output contract\n\nAll modes MUST emit `\u003Ctarget-dir>\u002FTHREAT_MODEL.md` conforming to `schema.md`\nin this directory. **Read `schema.md` immediately before you write the file**,\nnot at routing time; in interview mode the gap between routing and emit can be\nmany turns, and an early read will be evicted before it's used.\n\nAfter writing the file, print to the user:\n\n1. The path to `THREAT_MODEL.md`.\n2. The top 5 threats by likelihood × impact (id, one-line description, L×I).\n3. For `bootstrap`: any open questions the code could not answer (these seed a\n   later `interview` pass).\n4. For `interview`: any owner statements that could not be verified in code\n   (these seed follow-up code review).\n\n---\n\n## References\n\n- [docs\u002Fsecurity.md](..\u002F..\u002F..\u002Fdocs\u002Fsecurity.md) and\n  [docs\u002Fprompting.md](..\u002F..\u002F..\u002Fdocs\u002Fprompting.md) for the engagement-context\n  and authorization framing this skill inherits.\n","# threat-model\n\nA threat model answers **\"what could go wrong with this system, who would do\nit, and what should we do about it?\"** independently of whether any specific\nbug has been found yet. It is the map; vulnerability discovery is the metal\ndetector. A good threat model tells the pipeline where to look and tells triage\nwhich findings matter.\n\n**Litmus test:** If patching one line of code makes an entry disappear, it was\na vulnerability, not a threat. A threat (\"attacker achieves RCE via untrusted\nmedia parsing\") still stands after every known bug is fixed; a vulnerability\n(\"`dr_wav.h:412` doesn't bounds-check `chunk_size`\") does not. This skill\nproduces threats. Vulnerabilities appear only as **evidence** that raises a\nthreat's likelihood score.\n\n**Invocation:** `\u002Fthreat-model [bootstrap-then-interview|bootstrap|interview] \u003Ctarget-dir> [flags]`\n\n---\n\n## Step 0 — Safety preamble (always runs first)\n\nThis skill performs **static analysis only**. It reads source, git history,\nand any vulnerability reports the user supplies, and writes a single output\nfile (`\u003Ctarget-dir>\u002FTHREAT_MODEL.md`). It does not build, execute, fuzz, or\nmodify the target, and does not make network requests against the target's\ninfrastructure.\n\nBefore proceeding, confirm and state in your first response:\n\n1. The target directory exists and is a local checkout you can read.\n2. You will not execute any code from the target directory.\n3. If `--vulns` points at a URL or you are asked to \"fetch CVEs\", you will\n   query only public advisory databases (NVD, GitHub Security Advisories, the\n   project's own issue tracker) and never the target's live deployment.\n\nIf the user asks you to validate a threat by running an exploit, decline and\npoint them at the `vuln-pipeline` (README Step 2) instead.\n\n---\n\n## Step 1 — Route to a mode\n\nParse `$ARGUMENTS`:\n\n| First token | Route to |\n|---|---|\n| `interview` | Read `interview.md` in this directory and follow it. |\n| `bootstrap` | Read `bootstrap.md` in this directory and follow it. |\n| `bootstrap-then-interview` | Bootstrap first, then interview seeded from the draft. See below. |\n| anything else, or empty | Ask the user: **\"Is someone who owns or built this system available to answer questions in this session?\"** Yes and the codebase is checked out → recommend `bootstrap-then-interview`. Yes but no codebase → `interview.md`. No → `bootstrap.md`. |\n\nAll modes write the same artifact (`THREAT_MODEL.md`, schema in `schema.md`)\nso downstream consumers (pipeline `recon`\u002F`judge`, verifier agents) do not need\nto know which mode produced it.\n\n| | `interview` | `bootstrap` |\n|---|---|---|\n| **Needs** | An application owner present in the session | A local checkout; optionally past vulns |\n| **Method** | Four-question framework: conversational walk through *what are we working on → what can go wrong → what are we going to do about it → did we do a good job* | Five stages: parallel research swarm → synthesize sections 1-3 + vuln table → generalize vulns into threat classes → STRIDE gap-fill → emit |\n| **Best for** | New systems, design reviews, systems where the risk lives in business logic the code doesn't show | Inherited systems, third-party code, OSS dependencies, anything with a CVE history |\n| **Provenance tag** | `interview` | `bootstrap` |\n\n**Context durability.** Interview mode is multi-turn; tool results from early\nreads may be evicted before you need them. To stay resilient:\n\n- Do **not** read `interview.md` or `bootstrap.md` in full up front. Read the\n  mode file (or the relevant section of it) **at the point you need it**, one\n  question or stage at a time.\n- If a re-read via the Read tool is refused as \"file unchanged\", the prior\n  result was evicted; reload with `cat \u003Cpath>` via Bash instead.\n\n**Interview backbone** (so you can proceed even if `interview.md` is\nunavailable mid-session):\n\n| Q | Question | Fills schema sections |\n|---|---|---|\n| Q1 | What are we working on? | section 1 context, section 2 assets, section 3 entry points |\n| Q2 | What can go wrong? | section 4 threat rows (id, threat, actor, surface, asset) |\n| Q3 | What are we going to do about it? | section 4 impact\u002Flikelihood\u002Fstatus\u002Fcontrols; section 5 deprioritized; section 8 recommended mitigations |\n| Q4 | Did we do a good job? | validate ranking, coverage check, section 6 open questions |\n\n### `bootstrap-then-interview` mode\n\nWhen the owner is available *and* the codebase is checked out, this is the\nrecommended path: the owner's time goes to refining a code-grounded draft\ninstead of describing the system from scratch.\n\n1. Tell the owner: \"I'll read the code first and come back with a draft\n   (about 5-10 min), then we'll walk it together. Want that, or would you\n   rather start cold?\" Only proceed if they opt in; otherwise fall back to\n   `interview.md`.\n2. Read `bootstrap.md` and follow it end-to-end. Write\n   `\u003Ctarget-dir>\u002FTHREAT_MODEL.md`.\n3. Immediately continue into interview mode: read `interview.md` and follow\n   it with `--seed \u003Ctarget-dir>\u002FTHREAT_MODEL.md` in effect. The section 6 open\n   questions from bootstrap become your Q1-Q4 prompts; the owner confirms,\n   corrects, and adds rather than starting from nothing.\n4. Overwrite `\u003Ctarget-dir>\u002FTHREAT_MODEL.md` with the refined model. Set\n   provenance `mode: bootstrap-then-interview`.\n\nThe same flow is available manually: run `bootstrap` first, then\n`interview --seed \u003CTHREAT_MODEL.md>` in a later session.\n\n---\n\n## Step 2 — Shared output contract\n\nAll modes MUST emit `\u003Ctarget-dir>\u002FTHREAT_MODEL.md` conforming to `schema.md`\nin this directory. **Read `schema.md` immediately before you write the file**,\nnot at routing time; in interview mode the gap between routing and emit can be\nmany turns, and an early read will be evicted before it's used.\n\nAfter writing the file, print to the user:\n\n1. The path to `THREAT_MODEL.md`.\n2. The top 5 threats by likelihood × impact (id, one-line description, L×I).\n3. For `bootstrap`: any open questions the code could not answer (these seed a\n   later `interview` pass).\n4. For `interview`: any owner statements that could not be verified in code\n   (these seed follow-up code review).\n\n---\n\n## References\n\n- [docs\u002Fsecurity.md](..\u002F..\u002F..\u002Fdocs\u002Fsecurity.md) and\n  [docs\u002Fprompting.md](..\u002F..\u002F..\u002Fdocs\u002Fprompting.md) for the engagement-context\n  and authorization framing this skill inherits.\n",{"name":5,"description":7,"allowed-tools":24,"argument-hint":37},[25,26,27,28,29,30,31,32,33,34,35,36],"Read","Glob","Bash(python3 .claude\u002Fskills\u002F_lib\u002Fcheckpoint.py:*)","Grep","Write","Bash(git:*)","Bash(gh api:*)","Bash(find:*)","Bash(ls:*)","Bash(cat:*)","AskUserQuestion","Task","[bootstrap-then-interview|bootstrap|interview] \u003Ctarget-dir> [--vulns \u003Cfile>] [--design-doc \u003Cfile>] [--seed \u003CTHREAT_MODEL.md>] [--fresh]","main","d69096e239f2a89aabc2cd078c47664a9d3b1e11","04-安全渗透\u002Fthreat-model",3,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts","https:\u002F\u002Fgithub.com\u002Flza6","个人vx：Tf00798\r\n个人✈：@TFGY999","2026-10-08 14:29:22",{"source":8,"skillName":5,"commands":47,"prompt":57,"zipAvailable":58},[48,51,54],{"tool":49,"command":50},"npx","npx skills add lza6\u002FArmor-piercing-system-prompts --skill threat-model",{"tool":52,"command":53},"bunx","bunx skills add lza6\u002FArmor-piercing-system-prompts --skill threat-model",{"tool":55,"command":56},"pnpm","pnpm dlx skills add lza6\u002FArmor-piercing-system-prompts --skill threat-model","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fthreat-model\u002FSKILL.md",true,[60,66,70,74],{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"bootstrap.md","REFERENCE","参考文档",22018,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fthreat-model\u002Fbootstrap.md",{"path":67,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":68,"sourceHtmlUrl":69},"interview.md",9053,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fthreat-model\u002Finterview.md",{"path":71,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":72,"sourceHtmlUrl":73},"README.md",5187,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fthreat-model\u002FREADME.md",{"path":75,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":76,"sourceHtmlUrl":77},"schema.md",6844,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fthreat-model\u002Fschema.md",[79,87,92,99,106,113],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":86},"01M7FFG485KY83TD6HSC4YPCMC","obsidian-clipper-template-creator","Guide for creating templates for the Obsidian Web Clipper. Use when you want to create a new clipping template, understand available variables, or format clipped content.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator\u002FSKILL.md",[],[],"2026-11-16 10:05:34",{"skillNo":88,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":86},"01M7FFG0T6AZ0STEJPSB3CX84K","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator-antigravity-awesome-skills-main\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":86},"01M7FFFY0QEM14SKWPDYGMXM2P","python-api","Programmatic Object-Detection-Metrics API guidance for BoundingBox, BoundingBoxes, Evaluator, enums, coordinate conversion, AP\u002FmAP semantics, and source-style import diagnostics.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Fpython-api\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":86},"01M7FFFVZQZ2GJ0C0TGP7JHD20","file-evaluation","Folder-based PASCAL VOC AP\u002FmAP evaluation from ground-truth and detection text files.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Ffile-evaluation\u002FSKILL.md",[],[],{"skillNo":107,"name":108,"nameZh":6,"description":109,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":86},"01M7FFFSFNYSE77SJ6PE1CK2DA","object-detection-metrics","Guides legacy Object-Detection-Metrics PASCAL VOC AP\u002FmAP evaluation from detection text folders and direct Python API classes.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002FSKILL.md",[],[],{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":86},"01M7FFFP0GCK1RVXBCPQ8APBMT","nx-workspace-patterns","Configure and optimize Nx monorepo workspaces. Use when setting up Nx, configuring project boundaries, optimizing build caching, or implementing affected commands.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fnx-workspace-patterns\u002FSKILL.md",[],[]]