The skill exposes hardcoded credentials, lacks defined tool constraints, uses unpinned dependencies, and contains an insecure configuration system that permits arbitrary command injection.
npx skills add https://github.com/firecrawl/ai-research-skillsHardcoded credentials detected. Passwords, API keys, and secrets should not be hardcoded in source code. Use environment variables or a secrets manager instead.
api_key="nvapi-your-key-here"
The Python API example demonstrates passing API keys as plaintext strings in code, which risks accidental commit to version control systems.
api_key="nvapi-your-key-here"
The configuration system allows passing arbitrary environment variables and execution parameters to the evaluation harness. An attacker could potentially inject malicious commands or environment variables into the evaluation process if they control the config file. [ensemble: confirmed by 3/3 passes; severity set to the agreed median (ADR-0067).]
env_vars:
HF_TOKEN: HF_TOKENNER model detected organization in skill content (confidence: 0.52)
L*
NER model detected organization in skill content (confidence: 0.54)
N****
NER model detected organization in skill content (confidence: 0.57)
S********
NER model detected organization in skill content (confidence: 0.59)
D*****
NER model detected organization in skill content (confidence: 0.61)
E*********
NER model detected organization in skill content (confidence: 0.62)
M*******
NER model detected organization in skill content (confidence: 0.67)
M***
NER model detected organization in skill content (confidence: 0.69)
N******
NER model detected organization in skill content (confidence: 0.72)
S******
NER model detected organization in skill content (confidence: 0.75)
N*****
NER model detected organization in skill content (confidence: 0.81)
H********
NER model detected organization in skill content (confidence: 0.90)
O********
NER model detected organization in skill content (confidence: 0.97)
L*****
NER model detected organization in skill content (confidence: 0.57)
*
NER model detected organization in skill content (confidence: 0.79)
*
NER model detected organization in skill content (confidence: 0.81)
*
NER model detected organization in skill content (confidence: 1.00)
M**
Unpinned dependency installation — package installed without version pinning
pip install nemo-evaluator-launcher
SKILL.md links to "references/adapter-system.md" but the file is not part of the skill package — the workflow silently degrades or the content is sourced elsewhere at runtime
[references/adapter-system.md](references/adapter-system.md)
SKILL.md links to "references/configuration.md" but the file is not part of the skill package — the workflow silently degrades or the content is sourced elsewhere at runtime
[references/configuration.md](references/configuration.md)
SKILL.md links to "references/custom-benchmarks.md" but the file is not part of the skill package — the workflow silently degrades or the content is sourced elsewhere at runtime
[references/custom-benchmarks.md](references/custom-benchmarks.md)
SKILL.md links to "references/execution-backends.md" but the file is not part of the skill package — the workflow silently degrades or the content is sourced elsewhere at runtime
[references/execution-backends.md](references/execution-backends.md)
The SKILL.md 'name' field does not match the skill's parent directory name, which the Agent Skills spec requires. A mismatch can confuse skill resolution or mask the skill's real identity.
nemo-evaluator-sdk ≠ nemo-evaluator
[](https://mondoo.com/ai-agent-security/skills/github/firecrawl/ai-research-skills/nemo-evaluator)<a href="https://mondoo.com/ai-agent-security/skills/github/firecrawl/ai-research-skills/nemo-evaluator"><img src="https://mondoo.com/ai-agent-security/api/badge/github/firecrawl/ai-research-skills/nemo-evaluator.svg" alt="Mondoo Skill Check" /></a>https://mondoo.com/ai-agent-security/api/badge/github/firecrawl/ai-research-skills/nemo-evaluator.svgSkills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.