The Agent Skills Directory

[COMMAND_EXECUTION]: The validation script scripts/validate-handoff.mjs utilizes child_process.spawnSync to execute the tsx binary found in the project's local node_modules directory.
[REMOTE_CODE_EXECUTION]: The script scripts/validate-handoff.mjs performs dynamic code execution by using import() on computed file paths (dist/skills/handoff-validation.js or src/skills/handoff-validation.ts) to load its validation logic.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its data processing workflow.
Ingestion points: Untrusted data enters the agent context via the lore-recall tool and by reading JSON handoff files.
Boundary markers: No delimiters or instructions are used to distinguish or ignore embedded instructions in the ingested data.
Capability inventory: The skill has the capability to execute local scripts and spawn subprocesses via the validate-handoff.mjs script.
Sanitization: There is no evidence of validation or sanitization of external content before it is processed by the agent.

lore-handoff