The Agent Skills Directory

[DATA_EXFILTRATION]: The skill includes comprehensive guidance on preventing data leaks. In references/allure-evidence.md, there is a dedicated 'Sensitive Data And Redaction' section that instructs users to redact or omit secrets, API keys, tokens, and PII from test artifacts before they are attached to reports.
[COMMAND_EXECUTION]: The skill utilizes the allure CLI (specifically allure agent) via npx. This is the intended and standard method for running the Allure agent in modern Node.js environments. The commands are scoped to test execution and report generation.
[INDIRECT_PROMPT_INJECTION]: The skill involves processing external data from test runs, such as logs and manifests.
Ingestion points: The skill instructs the agent to review index.md, manifest/run.json, manifest/test-events.jsonl, manifest/tests.jsonl, and manifest/findings.jsonl (found in SKILL.md).
Boundary markers: No explicit boundary markers or 'ignore embedded instructions' warnings are defined for the parsing of these external files.
Capability inventory: The skill has the capability to execute shell commands using npx allure agent (found in SKILL.md).
Sanitization: The skill emphasizes secret redaction in references/allure-evidence.md but does not explicitly mention sanitizing processed data for prompt injection characters, though the risk is mitigated by the structured nature of Allure artifacts and standard LLM guardrails.
[SAFE]: The skill follows professional security best practices for reporting and testing. The external resources mentioned (Allure CLI) are well-known, industry-standard testing tools provided by the framework author.

allure-test-agent