Skills/npm/skill-architecture

skill-architecture@1.1.1

VERIFIED· install only

Publisher claims: "Enterprise-grade, local-first, zero-egress software architecture generator and multi-agent skill installer." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry npm (MCP server)Publisher fiky0911 · publishing for 9 daysInstalls 542Last detonated 8d ago
$claude mcp add skill-architecture -- npx -y skill-architecture
WatchReport an issue

Coverage

Install scripts ran; no agent task scenario for this bundle.
Tested
install scripts, with decoy credentials in place · second launch with the clock 45 days ahead · every file read, file written and host contacted by what ran
Not tested
use of the skill by an agent

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; install scripts ran; no agent task scenario for this bundle) on 2026-11-14 and scored it 80/100 (Verified). In what ran, nothing we could detect went wrong: no undeclared network egress, no credential reads, no persistent instruction writes. What it does with a real credential was not observed; the static findings and the publisher carry the rest of the score. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 1 s · 2 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessinstall script run in the sandbox
0.1sprocessinstall script run in the sandbox
0.2sprocessexec /usr/bin/dirname
0.2sprocessexited 1: Node.js v22.23.5
0.2sprocessscript backend runs setup only
0.3sprocessscript backend runs setup only
0.4sprocesssecond launch, clock +45 days, CI unset
0.5sprocessexited 1: Node.js v22.23.5
0.6sprocessexec /usr/bin/dirname
0.6sprocessexited 1: Node.js v22.23.5

Static findings

CRITICAL
Prompt injection attemptSKILL.md:171 · Prompt injection attempt — model control tokens- Ubiquitous: *The [system] shall [action].*Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Access to sensitive file pathsruntime-firewall/policy.yaml:13 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files- "~/.aws/**"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Fake instruction boundary tagadapters/claude/SKILL.md:48 · Fake instruction boundary tag<instructions>Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Access to sensitive file pathsadapters/claude/SKILL.md:16 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files2. Credential Protection: You must NEVER inspect, read, or print contents from `.env`, `*.pem`, `*.key`, `*.keystore`, `*.crt`, `secrets/`, `credentials/`, or authentication tokens.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathspolicies/filesystem-scope.json:31 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"~/.aws/**",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Prompt injection attemptSKILL.md:173 · Prompt injection attempt — model control tokens- State-driven: *While [in state], the [system] shall [action].*Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Access to sensitive file pathspolicies/filesystem-scope.json:30 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"~/.ssh/**",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsscripts/sanitize-input.js:12 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files/(^|[\\/])\.env($|\.)/i,Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsadapters/antigravity/SKILL.md:13 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files- **Zero Credential Access**: You must NEVER read, parse, display, or store secrets from `.env`, `.env.*`, `*.pem`, `*.key`, `*.keystore`, `*.crt`, or `secrets/`.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Prompt injection attemptSKILL.md:172 · Prompt injection attempt — model control tokens- Event-driven: *When [event], the [system] shall [action].*Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Access to sensitive file pathsruntime-firewall/policy.yaml:18 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files- "**/.env*"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Prompt injection attemptSKILL.md:175 · Prompt injection attempt — model control tokens- Optional: *Where [feature enabled], the [system] shall [action].*Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Access to sensitive file pathsSKILL.md:29 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files- Environment files: `.env`, `.env.*`, `*.env`Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathspolicies/filesystem-scope.json:18 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"**/.env.*",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsruntime-firewall/deny-rules.json:11 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"~/.ssh/**",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Fake instruction boundary tagadapters/claude/SKILL.md:77 · Fake instruction boundary tag</instructions>Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Access to sensitive file pathsSKILL.md:31 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files- Credential folders: `secrets/`, `credentials/`, `private/`, `~/.ssh/`, `~/.aws/`Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Prompt injection attemptSKILL.md:174 · Prompt injection attempt — model control tokens- Unwanted event: *If [error], then the [system] shall [action].*Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
CRITICAL
Prompt injectionSKILL.md:275 · Prompt injection — known jailbreak technique reference- Untuk Web UI: Ambil kode sumber komponen/tema nyata via UI MCP yang terpasang (`neobrutalism`, `shadcn-ui`, `magicui`, `ui-registry`, `tailwindcss`) dan catat di `reference-log.md`.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Access to sensitive file pathsruntime-firewall/deny-rules.json:12 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"~/.aws/**",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Prompt injectionSKILL.md:277 · Prompt injection — known jailbreak technique reference- Pilihan teknologi mobile sepenuhnya menyesuaikan preferensi dan stack yang diinginkan user (React Native/Expo, Flutter, SwiftUI, Jetpack Compose, Kotlin Multiplatform, dll.).Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Access to sensitive file pathsadapters/codex/SKILL.md:14 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files- **Zero Credential Access**: Never read, parse, or output values from `.env`, private keys (`*.pem`, `*.key`), certificates, or secret directories.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsruntime-firewall/deny-rules.json:5 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"**/.env*",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathspolicies/filesystem-scope.json:19 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"**/*.env",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Prompt injectionSKILL.md:274 · Prompt injection — known jailbreak technique reference…ck Compose, Tamagui, NativeWind), agent **WAJIB** memverifikasi sintaks, props, dan API resmi langsung dari dokumentasi terkini (via Context7 MCP untuk pustaka/SDK…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Access to sensitive file pathspolicies/filesystem-scope.json:17 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"**/.env",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsruntime-firewall/policy.yaml:14 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files- "~/.ssh/**"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
LOW
Prompt injection attemptreferences/requirements/srd.md:78 · Prompt injection attempt — model control tokens · informational: not an instruction file · informational: not an instruction file- Syntax: `If [error/fault condition], then the [system] shall [action].`Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
npx invocation of non-well-known packagescripts/install.js:588 · npx invocation of non-well-known package — may execute untrusted code$ npx skill-architecture --global --target antigravityFix: Pin versions and sources; vendor anything fetched at install time.
AST06
LOW
Rule modification attemptreferences/architecture/fsd.md:111 · Rule modification attempt — overwriting existing instructions · informational: not an instruction file · informational: not an instruction file### Rule 2: Strict Cross-Slice Import IsolationFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Access to sensitive file pathsskill.yaml:94 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files- "**/.env*"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
LOW
Authorization escalation languageadapters/README.md:99 · Authorization escalation language · informational: not an instruction file · informational: not an instruction file- Windows requires Administrator privileges or Developer Mode to create symlinks. Use `--mode copy` on Windows environments:Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
LOW
Prompt injectiondocs/architecture/README.md:3 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction fileDokumen ini menjelaskan keputusan arsitektural utama proyek, termasuk diagram C4 dan pemetaan modul.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Fake instruction boundary tagadapters/README.md:65 · Fake instruction boundary tag · informational: not an instruction file · informational: not an instruction file- **Syntax**: Anthropic Claude structured XML blocks: `<role>`, `<instructions>`, `<guidelines>`, `<capabilities>`, `<command_routing>`.Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
Access to sensitive file pathsdocs/spec/README.id.md:28 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files…dungan Kredensial Absolut**: Skill ini secara struktural dilarang membaca file `.env`, sertifikat, kunci privat, atau token autentikasi agen. Pemindaian rahasia baw…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
npx invocation of non-well-known packagescripts/install.js:590 · npx invocation of non-well-known package — may execute untrusted code$ npx skill-architecture --global --dry-runFix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
Access to sensitive file pathsskill.yaml:101 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files- "~/.ssh/**"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
LOW
Prompt injectionadapters/README.md:99 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file- Windows requires Administrator privileges or Developer Mode to create symlinks. Use `--mode copy` on Windows environments:Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
Access to sensitive file pathsdocs/spec/TEST_INFRA.md:20 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files…usively synthetic, dummy credentials. Tests verify that real host credentials (`~/.ssh`, `~/.aws`, `.env`, API keys) are strictly shielded and unreadable by the skill…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
LOW
Prompt injectiondocs/workflow.md:11 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file- **Arsitektur** – menampilkan keputusan arsitektural dan diagram C4.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectiondocs/spec/README.id.md:28 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file…ertifikat, kunci privat, atau token autentikasi agen. Pemindaian rahasia bawaan dan penyamaran otomatis menetralisir kredensial yang tersesat sebelum operasi penul…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectiondocs/workflow.md:3 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction fileDokumen ini menjelaskan alur kerja proyek secara **high‑level** dan menghubungkan semua bagian dokumentasi yang ada.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectiondocs/workflow.md:14 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file- **Desain Sistem** – pola, komponen, dan strategi implementasi.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injection attemptreferences/requirements/srd.md:68 · Prompt injection attempt — model control tokens · informational: not an instruction file · informational: not an instruction file- Syntax: `When [trigger/event], the [system] shall [action].`Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectiondocs/requirements/README.md:3 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction fileDokumen ini berisi **Business Requirement Document (BRD)**, **Product Requirement Document (PRD)**, dan **System Requirement Document (SRD)** untuk proyek.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
npx invocation of non-well-known packagescripts/install.js:589 · npx invocation of non-well-known package — may execute untrusted code$ npx skill-architecture --local --agent claudeFix: Pin versions and sources; vendor anything fetched at install time.
AST06
LOW
Prompt injectiondocs/workflow.md:13 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file- **Requirement** – Business Requirement Document (BRD), Product Requirement Document (PRD), dan System Requirement Document (SRD).Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injection attemptreferences/requirements/srd.md:83 · Prompt injection attempt — model control tokens · informational: not an instruction file · informational: not an instruction file- Syntax: `Where [feature flag/option enabled], the [system] shall [action].`Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectiondocs/architecture/README.md:7 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file- **Diagram C4** – diagram Mermaid berada di `assets/uml/` dan dapat direferensikan di sini.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
npx invocation of non-well-known packagescripts/install.js:586 · npx invocation of non-well-known package — may execute untrusted code$ npx skill-architecture --globalFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Access to sensitive file pathsdocs/spec/TEST_INFRA.md:117 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files…ostics, fallback to clean initialization | | **Path Traversal Attacks** | `../../etc/passwd`, `..\\..\\windows\\system32`, `docs/../../malicious.md`, null byte `%00` | Har…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
LOW
Prompt injectiondocs/spec/README.id.md:23 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file…ntesis spesifikasi kelas enterprise, skema data, pengaman sistem terdistribusi, dan token UI visual ke dalam hierarki `./docs/` yang terorganisir di dalam proyek A…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
npx invocation of non-well-known packagescripts/install.js:587 · npx invocation of non-well-known package — may execute untrusted code$ npx skill-architecture --global --agent cursorFix: Pin versions and sources; vendor anything fetched at install time.
AST06
LOW
Rule modification attemptreferences/architecture/fsd.md:104 · Rule modification attempt — overwriting existing instructions · informational: not an instruction file · informational: not an instruction file### Rule 1: Strict Unidirectional Downward ImportsFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
LOW
Prompt injectiondocs/spec/README.id.md:10 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file> Generator arsitektur perangkat lunak dan kerangka spesifikasi kelas enterprise, berbasis lokal, zero-egress untuk agen AI.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectiondocs/spec/README.id.md:21 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file… IDE, Continue.dev, Codeium Windsurf, Cline Autonomous Agent, OpenAI Codex CLI, dan Google Antigravity. Tidak seperti prompt AI biasa yang menghasilkan saran kode…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectiondocs/workflow.md:15 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file- **UI** – komponen UI, tema, dan panduan styling.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectiondocs/spec/README.id.md:26 · Prompt injection — known jailbreak technique reference · informational: not an instruction file · informational: not an instruction file…i Logika atas Heuristik**: Semua paradigma arsitektur, trade-off desain sistem, dan klausul kebutuhan berakar pada standar formal (ISO/IEC/IEEE 29148:2018, BABOK G…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injection attemptreferences/requirements/srd.md:73 · Prompt injection attempt — model control tokens · informational: not an instruction file · informational: not an instruction file- Syntax: `While [in state], the [system] shall [action].`Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Access to sensitive file pathsskill.yaml:102 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files- "~/.aws/**"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
LOW
Prompt injection attemptreferences/requirements/srd.md:63 · Prompt injection attempt — model control tokens · informational: not an instruction file · informational: not an instruction file- Syntax: `The [system] shall [action].`Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01

Cost to the agent

Instructions loaded
≈8,820 tok
4 files · 80 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

No advisories in OSV for skill-architecture@1.1.1 as of 4 h ago. This is the package itself, not its dependencies.