speckit-converge

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill instructs the agent to read and then echo and execute hook command strings from .specify/extensions.yml (and to consider raw $ARGUMENTS), so any secrets contained in those inputs would be emitted verbatim (EXECUTE_COMMAND / Command / Prompt), creating an exfiltration risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The spec requires running repository-provided scripts and invoking arbitrary extension hooks from .specify/extensions.yml (including mandatory hooks), which enables execution of untrusted code and poses high risk of backdoor/RCE and supply-chain abuse.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The prompt requires the agent to execute repository scripts and to actually invoke mandatory extension hooks (pre- and post-converge) as real commands, which allows running arbitrary commands that can modify the machine state or escalate privileges.