The Agent Skills Directory

[CREDENTIALS_UNSAFE]: Accesses the local configuration file ~/.qodo/config.json to retrieve the API_KEY for authentication.\n
Evidence: API_KEY=$(python3 -c "import json,os; c=json.load(open(os.path.expanduser('~/.qodo/config.json'))); print(c['API_KEY'])") (SKILL.md).\n- [EXTERNAL_DOWNLOADS]: Fetches coding rules from official Qodo platform API endpoints.\n
Evidence: Uses curl to request rules from https://qodo-platform.qodo.ai/rules/v1 or configured environment subdomains (pagination.md).\n- [COMMAND_EXECUTION]: Uses Bash to execute system utilities for repository inspection and configuration parsing.\n
Evidence: Executes git origin, python3 -c, uuidgen, and curl (SKILL.md, pagination.md).\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by incorporating externally fetched rule descriptions into the agent context.\n
Ingestion points: Rules are retrieved from the Qodo platform API (pagination.md).\n
Boundary markers: Loaded rules are delimited by a "📋 Qodo Rules Loaded" header and a "---" footer (output-format.md).\n
Capability inventory: The skill can execute shell commands via Bash, including git, curl, and python3 (SKILL.md).\n
Sanitization: Rule descriptions fetched from the remote API are not explicitly sanitized before being rendered into the prompt context.

qodo-get-rules