The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill's documentation provides instructions for installing the ripwire utility using a shell script hosted on an official Red Hat Emerging Technologies GitHub repository. This is a documented prerequisite for the skill's route discovery functionality.
[COMMAND_EXECUTION]: The Python scripts included with the skill (build_filtered_oas.py and parse_pact_coverage.py) utilize subprocess.run to execute external command-line tools, specifically ripwire and pact-broker. These executions are intended for scanning consumer codebases and managing Pact files during the coverage analysis process.
[INDIRECT_PROMPT_INJECTION]: The skill processes external, potentially untrusted data which creates a surface for indirect prompt injection.
Ingestion points: The skill reads OpenAPI specifications and Pact JSON files from local storage, URLs, or Pact brokers, and it performs text-based scans of consumer source code files.
Boundary markers: The SKILL.md instructions explicitly mandate that the agent must validate all OpenAPI and Pact files, prohibits the hallucination of spec content, and requires the agent to obtain user confirmation before selecting specific files for analysis.
Capability inventory: The skill performs network requests via urllib to fetch pacts from brokers and executes local subprocesses to run analysis tools.
Sanitization: The skill employs yaml.safe_load and json.loads to parse structured data files, and it utilizes jsonref for secure reference resolution within OpenAPI specs.

pact-coverage