The Agent Skills Directory

Data Exposure & Exfiltration (MEDIUM): The diagnostic script scripts/check-env.sh accesses sensitive environment variables and configuration files.
Evidence: The script reads .env, .env.local, .env.development, and .env.production files using grep to check for variables.
Evidence: It uses printenv to retrieve values for RECUR_SECRET_KEY and RECUR_WEBHOOK_SECRET, displaying the first 12 characters of these secrets.
Risk: While the script includes partial masking, it exposes sensitive file contents and partial secret keys to the agent's runtime environment, which could be exploited if the agent's context is compromised.
Indirect Prompt Injection (LOW): The skill ingests untrusted data from local configuration files which could influence agent behavior.
Ingestion points: package.json and .env files are parsed via scripts/check-env.sh.
Boundary markers: Absent.
Capability inventory: Shell script execution and environment variable access.
Sanitization: None.
External Downloads (LOW): The skill recommends installing the recur-tw package from npm.
Source: npmjs.com (Standard registry; no suspicious behavior identified in the instruction).

recur-quickstart