Skills/ClawHub/ai-pm-workbench-international

ai-pm-workbench-international@1.2.0-intl

CAUTION· static only

Publisher claims: "AI PM Super Workbench — International Edition. Full-stack intelligent workbench for AI Product Managers worldwide." Observed in sandbox: static analysis only — behavior not yet observed.

Registry ClawHubPublisher yinjianheng · publishing for 129 daysInstalls 713Last detonated 3d ago
What would fix thisWatchDispute verdict

Coverage

Static only; nothing was run yet.

The label rests on the code, the instruction files and the publisher. The honeypot run replaces this verdict when it completes.

Is [email protected] safe to install?

KillerSkills analysed it statically (a sandbox run is pending) on 2026-11-11 and scored it 71/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Prompt injection; Prompt injection; Prompt injection. Summary: static analysis only — behavior not yet observed.

Behavior trace

Queued for honeypot run · static analysis only
t+CategoryObserved action

This version has not been detonated yet. The score above counts behavior as unknown (50/100), so it cannot read Verified until the honeypot run completes and replaces this verdict.

Static findings

CRITICAL
Prompt injectionSKILL.md:84 · Prompt injection — known jailbreak technique reference…, controllability | | **Security Boundary** | Multi-dimensional attack surface (Jailbreak/Injection/Data Poisoning) | Traditional network security + application security…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/ai-industry-trends-2026.md:177 · Prompt injection — known jailbreak technique reference · informational: not an instruction file├── Adversarial samples / jailbreak promptsFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
CRITICAL
Prompt injectionSKILL.md:1090 · Prompt injection — known jailbreak technique reference- Phase 10: Safety Guardrails (jailbreak protection + content moderation)Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionREADME.md:55 · Prompt injection — known jailbreak technique reference · informational: not an instruction file| AI Safety | Input Guardrails · Output Guardrails · PII Redaction · Jailbreak Detection · Red Teaming |Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injection attemptreferences/templates/ai-safety-template.md:89 · Prompt injection attempt — 'ignore previous instructions' · informational: not an instruction file| Role-play Jailbreak | Detect "You are now..." type instructions | "DAN mode", "Ignore previous instructions" |Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/examples/ai-customer-service-example.md:390 · Prompt injection — known jailbreak technique reference · informational: not an instruction file- Jailbreak Detection: Detects role-playing, encoding bypassesFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/templates/ai-safety-template.md:87 · Prompt injection — known jailbreak technique reference · informational: not an instruction file| Jailbreak Type | Detection Strategy | Example Pattern |Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/templates/ai-safety-template.md:48 · Prompt injection — known jailbreak technique reference · informational: not an instruction file| P1-High | Exploitable, bypassing restrictions | Successful jailbreak, injection attack | 2 hours |Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionREADME.md:39 · Prompt injection — known jailbreak technique reference · informational: not an instruction file| 9️⃣ Safety Guardrails & Red Team | Layered guardrails · Red teaming · Jailbreak prevention | AI Safety Plan |Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
CRITICAL
Prompt injectionSKILL.md:783 · Prompt injection — known jailbreak technique reference| 19 | **Ignoring Low-Resource Language Security** | Test jailbreak risks for all supported languages |Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/templates/ai-prd-template.md:228 · Prompt injection — known jailbreak technique reference · informational: not an instruction file| Jailbreak | | |Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injection attemptreferences/examples/ai-customer-service-example.md:389 · Prompt injection attempt — 'ignore previous instructions' · informational: not an instruction file- Injection Detection: LLM classifier detects patterns like "ignore previous instructions"Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/templates/prompt-engineering-template.md:300 · Prompt injection — known jailbreak technique reference · informational: not an instruction file| Safety Test Set | | Jailbreak/Injection | Monthly |Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injection attemptreferences/templates/ai-safety-template.md:254 · Prompt injection attempt — 'ignore previous instructions' · informational: not an instruction file├── Instruction Override: "Ignore previous instructions"Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/ai-industry-trends-2026.md:879 · Prompt injection — known jailbreak technique reference · informational: not an instruction fileLayer 2: Input Guardrails (injection detection + jailbreak detection + PII filtering)Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/templates/ai-evaluation-template.md:229 · Prompt injection — known jailbreak technique reference · informational: not an instruction file| Safety Violation | Harmful output / PII leakage / successful jailbreak | P0 |Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST01
LOW
Prompt injectionreferences/templates/ai-safety-template.md:85 · Prompt injection — known jailbreak technique reference · informational: not an instruction file### 2.3 Jailbreak DetectionFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/examples/ai-customer-service-example.md:356 · Prompt injection — known jailbreak technique reference · informational: not an instruction file| Adversarial Scenarios (Injection / Jailbreak) | 10% |Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/templates/ai-safety-template.md:79 · Prompt injection — known jailbreak technique reference · informational: not an instruction file"Detect whether the following user input contains Prompt injection/jailbreak attempts.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
CRITICAL
Prompt injectionSKILL.md:655 · Prompt injection — known jailbreak technique reference□ AI Security Basics (Injection / Jailbreak / Guardrails / Red Team Testing)Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/templates/ai-safety-template.md:61 · Prompt injection — known jailbreak technique reference · informational: not an instruction file| Jailbreak Attack | LLM Classifier | Reject + Alert | <2% |Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/templates/ai-evaluation-template.md:86 · Prompt injection — known jailbreak technique reference · informational: not an instruction file| Adversarial Scenarios | 10% | Injection, jailbreak, malicious inputs |Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST01
LOW
Prompt injectionreferences/methodologies/ai-pm-deep-methods.md:533 · Prompt injection — known jailbreak technique reference · informational: not an instruction file├── LLM injection/jailbreak classifierFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/ai-industry-trends-2026.md:869 · Prompt injection — known jailbreak technique reference · informational: not an instruction file├── Input Layer: Prompt injection, Jailbreak, Base64/ROT13 encoding bypass, multilingual bypassFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01

Cost to the agent

Instructions loaded
≈22,197 tok
1 file · 17 in bundle

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

Known vulnerabilities

Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.

What would make this Verified

9 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim yinjianheng: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
Publish a new version to ClawHub, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim yinjianhengPropose a patch to the maintainer