Validate all cross-links in the CLI Agent Spec specification — broken file links, missing schema sections, and schema↔requirement symmetry. Use when files have been added or edited, or to check the project is internally consistent.
Guide implementing the CLI Agent Spec specification in a CLI framework or tool. Audits command naming against Unix conventions, walks through requirements tier-by-tier (REQ-F → REQ-C → REQ-O), generates language-specific schema types, and verifies acceptance criteria. Use when building or extending a CLI framework to support AI agent orchestration.
Classify a failed agent CLI tool call against the CLI Agent Spec §N failure taxonomy. Given a failed command with stdout, stderr, and exit code, identifies the matching failure mode, returns an actionable workaround, and produces a memory string and skill patch to prevent recurrence. Use when a CLI invocation fails and you need to understand why and how to work around it.
Autonomous end-to-end CLI audit pipeline. Downloads and installs the CLI non-interactively, onboards it, scores proactive readiness, evaluates all Critical failure modes, and generates the full report bundle (dev, agent-dev, runtime, issues, index, LinkedIn post). Single command, zero human steps required in the happy path.
Generate a perspective-specific report from existing CLI evaluation findings. Five modes — dev (fix list for CLI authors), agent-dev (integration guide for agent builders), runtime (operational brief for AI agents), issues (concrete bugs and gaps an agent will hit), all (runs all four, saves to files, generates an index and a LinkedIn post). Requires findings from cli-agent-evaluate or cli-agent-evaluate-batch.
Evaluate a CLI tool against multiple CLI Agent Spec failure modes in one run. Accepts a severity filter, part number, or an explicit list of §N identifiers. Skips already-evaluated failure modes (resumable). Emits a scorecard table when done.
Evaluate a CLI tool against a single CLI Agent Spec failure mode. Runs the failure mode's check, scores 0–3, and provides an applicable agent workaround if the score is below 3. Use this for targeted single-failure-mode evaluation. For multi-failure-mode evaluation use cli-agent-evaluate-batch.
Onboard a CLI tool for agent evaluation — reads agent docs, detects OS, runtime, toolchain, and binary, discovers non-interactive flags and config. Saves the result as the CLI environment profile for reuse by cli-agent-evaluate and other skills. Run once per CLI before starting evaluations.