The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local shell script scripts/estimate-tokens.sh using the Bash tool to perform token and line counting for the project being audited.\n- [DYNAMIC_EXECUTION]: The skill uses the Workflow tool to execute workflows/audit-full-mapreduce.js, which dynamically manages sub-agents and synthesis logic for large-scale audits.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (entire codebases) which serves as an injection surface for malicious instructions hidden in the audited code.\n
Ingestion points: The Read tool is used in SKILL.md and workflows/audit-full-mapreduce.js to load all source files from the project directory into the agent's context.\n
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the audited files are provided in the skill's prompts.\n
Capability inventory: The skill utilizes powerful tools including Bash, Workflow, Agent, TaskCreate, PushNotification, and Read across its various phases.\n
Sanitization: The skill does not perform any sanitization, filtering, or escaping of the content read from files before it is processed by the LLM.\n- [DATA_EXFILTRATION]: The PushNotification tool is used to send audit summaries (finding counts and report locations) to an external notification service upon completion.

audit-full