The Agent Skills Directory

[PROMPT_INJECTION]: No evidence of malicious behavior overrides or safety bypass instructions. All instructions are focused on improving the quality and clarity of technical communication.
[DATA_EXFILTRATION]: No sensitive file access or unauthorized network operations were detected. The skill operates on technical documentation files provided through standard agent tasks.
[REMOTE_CODE_EXECUTION]: The skill does not define external dependencies or execute remote scripts. Code snippets provided are illustrative of internal framework usage (Task, FleetManager).
[COMMAND_EXECUTION]: No direct system command execution or shell spawning was found. The skill uses abstract task definitions for agent integration.
[INDIRECT_PROMPT_INJECTION]: The skill identifies documentation files and source code as input sources for generation and review tasks.
Ingestion points: Files like PaymentService.ts, README.md, and docs/api.md (via Agent Integration examples).
Boundary markers: Not explicitly defined in the provided snippets.
Capability inventory: Limited to document generation and review via qe-quality-analyzer and qe-api-contract-validator agents.
Sanitization: Standard for documentation processing tools; no explicit sanitization logic shown in the markdown templates.

technical-writing