The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run shell commands to validate sequence files using the vendor's synapseq tool (e.g., bin/synapseq -test and go run ./cmd/synapseq -test). These commands are scoped to validation and use the -test flag to avoid side effects.
[INDIRECT_PROMPT_INJECTION]: The skill processes external .spsq and .spsc files, which constitutes a potential surface for indirect prompt injection.
Ingestion points: Loading sequence content and dependencies as described in the sequence walkthrough section of SKILL.md.
Boundary markers: Detailed instructions guide the agent to interpret files strictly as audio scores and maintain grounded, behavioral descriptions.
Capability inventory: Shell command execution is available through the local synapseq validation binary.
Sanitization: The agent is instructed to treat the official parser and sequence builder as authoritative for validation rather than performing manual text interpretation.

explain-spsq