The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs shell command execution using the faf CLI tool (e.g., faf wjttc, faf score) to audit test tiers and calculate deterministic scores.
[PERSISTENCE]: The instruction provides a command (faf taf setup --write) that modifies the project's file system to create or update GitHub Action workflows in .github/workflows/taf.yml. This establishes a persistent automated execution mechanism within the repository's CI/CD environment.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze potentially untrusted data from test failure logs and bug reports to perform root-cause analysis and file WJTTC reports. This content could be crafted to include hidden instructions that influence the agent's behavior during analysis.
Ingestion points: Processes external test execution outputs and user-provided bug descriptions (documented in SKILL.md under the Reproduction and Reporting sections).
Boundary markers: There are no specific delimiters or instructions defined to ignore embedded prompts in the ingested data.
Capability inventory: The skill possesses shell execution capabilities and file-writing permissions within the project directory.
Sanitization: No sanitization or validation logic is specified for the content of processed logs or reports.

wjttc-tester