Caveat verdict
tilt-protocol
Accesses credentials AND makes external network calls
Before starting any session, the skill instructs fetching its own replacement from an external Railway.app URL and reviewing it for changes, creating a living-skill pattern where a third-party server can inject new instructions into every agent session without user review.
Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
What it does
These are capability combinations: each listed behavior occurs in the skill, but Caveat detects co-occurrence — it does not verify that one flows into another. Read the code to confirm a live chain.
Accesses credentials AND makes external network calls — potential credential theft
LLM02 · ASI03
Accesses credentials AND encodes data — may obfuscate stolen credentials
LLM02 · ASI03 · ASI04
Permission integrity
network_out
credential_access
package_install
Findings (4)
Pipe-to-shell pattern (curl | sh) — supply chain attack vector
SKILL.md · code · curl -L https://foundry.paradigm.xyz | bash
Pipe to bash — executes piped content as shell commands
SKILL.md · code · | bash
apt-get install — installs system packages
SKILL.md · code · apt-get install
Data URI with base64 payload — may embed malicious content
SKILL.md · code · data:application/json;base64,
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions — minimal attack surface.
network_outpackage_installcredential_accessdata_encoding Thanks — recorded.