The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Node.js and Bash scripts for identity reporting and data retrieval. It specifically invokes node scripts/identity-report.mjs during initialization to verify the tenant and host environment. It also utilizes standard CLI tools including linearis, gh, git, and sqlite3 to interact with the project management environment.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data by reading and acting upon human comments and requests from Linear tickets and status boards.
Ingestion points: Human comments on tickets and requests for project scaffolding are processed by the agent (identified in SKILL.md loop and references/scaffold.md).
Boundary markers: The instructions do not define specific delimiters or warnings to ignore potentially malicious instructions embedded in the user-provided text.
Capability inventory: The skill has the capability to write to local files (handoffs, event logs), interact with the Linear API via CLI, and launch new agent roles via a supervisor.
Sanitization: The skill uses jq within its Bash helpers (scripts/lib/canonical-event.sh) to ensure that telemetry data is safely formatted as JSON.

concierge