The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local script named db.mjs to manage run records and concurrency leases. This utility is used for operational logging and ensuring tasks do not run simultaneously.
[INDIRECT_PROMPT_INJECTION]: The skill acts as an ingestion point for external data through the refresh-backlog step, which processes 'new inbound' information.
Ingestion points: New inbound data is processed in SKILL.md via the refresh-backlog invocation.
Boundary markers: The skill does not define specific delimiters or instructions to prevent the agent from following directives embedded within the ingested data.
Capability inventory: The agent can execute the db.mjs script and trigger follow-up actions like act-post-decision as described in SKILL.md.
Sanitization: There is no evidence of data sanitization or validation performed on the external content before it is passed to subsequent steps.

go