ClawAudit verdict

openai-agent-sdk

88
🟢 Trusted
Low risk — reviewed by ClawAudit, behavior matches stated purpose

Receives external input AND uses eval

OpenAI Agents SDK tutorial skill for building multi-agent systems; dynamic_eval is inherent to the SDK's function_tool decorator pattern, and all examples are standard SDK usage with no malicious patterns.

Automated static analysis — not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.

0
security
40
transparency
70
maintenance

What it does

These are capability combinations: each listed behavior occurs in the skill, but ClawAudit detects co-occurrence — it does not verify that one flows into another. Read the code to confirm a live chain.

Capability combination critical

Receives external input AND uses eval — the remote code-injection pattern (data-flow not verified)

LLM01 · LLM05 · ASI01 · ASI05

Permission integrity

Code accesses API keys/tokens but declares no environment variables

credential_access

Installs packages at runtime — transitive dependencies are not auditable

package_install

Findings (2)

Pattern match critical

Possible hardcoded credential

SKILL.md · code · API_KEY="sk-your-api-key-here

Pattern match critical

Uses eval() — can execute arbitrary code

SKILL.md · code · eval(

Permissions & capabilities

No declared permissions — minimal attack surface.

package_installcredential_accessnetwork_indynamic_eval
Check another skill Browse the registry Auditing your own skills or configs? Use the API