talent-de-cv
3 findings (1 high, 2 medium), mainly data exfiltration, social engineering, prompt injection. Most recent scan 2026-02-06.
This skill presents MODERATE to HIGH security risks primarily related to data exfiltration, privacy violations, and transitive trust abuse. The core functionality involves collecting personal information (names, emails, job titles) and transmitting it to an external third-party service (talent.de) to create permanent public profiles. Key concerns include: (1) unvalidated network calls to external APIs with user data, (2) misleading claims about 'no signup required' when Access-ID is needed for advanced features, (3) lack of privacy controls or informed consent mechanisms, (4) transitive trust to external unverified content from talent.de URLs, and (5) encouraging AI agents to create potentially deceptive professional profiles. The skill lacks basic security controls like input validation, SSL verification, content integrity checking, or data protection measures. While there are no script files to analyze for code injection risks, the instruction-based data collection and external transmission patterns pose significant privacy and security risks. The missing optional metadata (license, compatibility, allowed-tools) is a minor issue but reduces transparency.
Scanner's overall assessment (LLM analyzer)
Second scanner: NVIDIA SkillSpector
Independent static run, no language model, over the same files as the Cisco scan above. How the two scanners compare across the whole corpus: Two scanners, one corpus.
What this skill does
Create free interactive digital identities and CVs on talent.de — 22 templates, permanent URL. No signup, no payment, no API key needed for basic use.
- Package
- 2 files, 28 KB
- SKILL.md
- 3838 words, 12+ sections
- Code
- None. Instructions only.
- Requires
- network:
True - Version
- 2.2.0
- Homepage
- https://www.talent.de
SKILL.md covers: talent.de — Digital Identity & CV Skill · Two Ways to Create · Skill Files · Instructions · Step 1: Decide Who This Is For · Step 2: Choose a Slug (Your URL) · Step 3: Choose a Template · Step 4: Build Your cv_data Object · Step 5: Send the API Request · Done! · Set Up Your Heartbeat · CV Data Reference
Findings
HIGH Unvalidated External Network Calls to Third-Party API Data Exfiltrationllm
The skill makes HTTP requests to an external domain (talent.de) without any validation, authentication, or security controls. The API endpoint https://www.talent.de/api/public/slugs/check accepts user-provided parameters (slug, firstName, lastName) that could be used to exfiltrate data or fingerprint users. The skill encourages creating permanent public profiles with personal information (names, emails, job titles) on an external service without informed consent or privacy controls.
SKILL.md
GET https://www.talent.de/api/public/slugs/check?slug=007&firstName=Max&lastName=Mustermann - User data (firstName, lastName) sent to external API without validation or consent mechanismMEDIUM Deceptive Marketing Claims - 'No Signup Required' Contradicts Access-ID Requirement Social Engineeringllm
The skill description claims 'No signup, no payment, no API key needed for basic use' but the instructions reveal that Way B (custom templates) requires an 'Access-ID' with a limit of '10 templates/day'. This is misleading social engineering that downplays authentication requirements. The distinction between 'basic use' (3 CVs/day) and advanced use (requiring Access-ID) is buried in a comparison table rather than prominently disclosed.
SKILL.md
Description: 'No signup, no payment, no API key needed for basic use' vs Table: 'Access-ID: Required (10 templates/day)' for Way BMEDIUM Transitive Trust to External Unverified Content Prompt Injectionllm
The skill instructs agents to read and follow instructions from external URLs (https://www.talent.de/skill.md, https://www.talent.de/heartbeat.md, https://www.talent.de/skills/template-create.md) without any content verification, integrity checking, or security validation. An attacker who compromises talent.de could inject malicious instructions that the agent would follow. The curl commands download and execute content from external sources without signature verification.
SKILL.md
Instructions to 'curl -s https://www.talent.de/skill.md' and 'Or just read them from the URLs above!' - delegating trust to external, unverified web contentWhat is in the package
| File | Kind | Size |
|---|---|---|
heartbeat.md | doc | 2.8 KB |
skill.md | doc | 25 KB |
External hosts referenced
www.talent.de www.moltbook.com github.com linkedin.com maxmustermann.dev
Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.
Environment variables referenced
INVALID_ACCESS_ID
Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.
What we checked
- Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
- Behavioral analyzer. Checks what the skill asks the agent to do against what its description claims it does.
- Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
- Trigger analyzer. Looks for instructions that fire on conditions the user did not ask for.
- Meta analyzer. A second model pass that validates, deduplicates and re-scores the other analyzers' findings.
What we did not check
- Runtime behaviour. The skill was not executed; findings come from reading its files.
- Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
- Content the analyzers cannot read: images, binaries, and some non-English text.
- Known scanner bypass classes. See the methodology and limitations page.
Sources
- Repository: https://clawhub.ai/skills/id-cv-resume-creator
- Raw scan result: digital-identity,-cv-&-resume-creator-scan.json
- Badge for your README:
- Request a re-scan