Coordinates agent and skill creation, testing, calibration, and iteration via plan files and scripts.
npx skills add https://github.com/slowman2084/meta-agent --skill meta-plan
Please help me install this Agent Skill. Skill: meta-plan Source: https://github.com/slowman2084/meta-agent/tree/main/source/skills/meta-plan Command: npx skills add https://github.com/slowman2084/meta-agent --skill meta-plan
Meta-plan solves the fragmentation and manual coordination involved in creating, testing, calibrating, and iterating Agents and Skills by providing a single, disk-backed planning and execution controller that supports resumable runs and artifact-driven state.
Use meta-plan to create an agent named 'lyrics-bot' and execute a baseline test run to generate a full evaluation report.
High-intent search queries and answers about installing and using this skill.
Agent lifecycle automation uses a disk-backed planning controller to generate structured plan markdown files, orchestrating creation, testing, calibration, and iteration across your repository workspace. It produces timestamped plans and manages tmp and bak artifacts for resumable execution.
Yes, resumable execution recovers context via scripts/context_tool.py and marks step completion based on files in the tmp/ directory. It enforces backups to bak/ and produces status.json for synchronization, allowing interrupted agent calibration workflows to resume accurately.
Generating test cases and rubrics for automated agent evaluation involves orchestrating meta-testcase-gen and meta-rubric-gen skills. The controller exports test inputs, executes cases across platform modes, and evaluates results using meta-eval-judge to produce a consolidated report.
Orchestrating multiple meta-* skills for prompt iteration requires a single planning controller that drives meta-ideal-state, meta-prompt-engineer, meta-debug, and meta-iterate. It manages artifact-driven state and produces evaluation reports to guide prompt refinement across the repository workspace.
Agent orchestration supports platform modes for baseline execution runs by exporting test inputs and evaluating results consistently. The controller applies platform-specific configurations during execution and consolidates outputs into a unified evaluation report with status synchronization.
Agent testing workflow synchronization fails when tmp/ artifacts are missing or scripts/status_tool.py cannot read status.json. The controller relies on these disk-backed artifacts to mark step completion, enforce bak/ backups, and recover context via scripts/context_tool.py for accurate state tracking.
riftzen-bitImplement actor-based thread-safe persistence with in-memory cache and atomic file writes.
slowman2084Analyze prompt histories to detect anti-patterns and generate retrospective reports.
fockusCoordinate three judge agents in iterative debates to produce consensus evaluations.
zhangzhang-111-iSynchronize Playwright test cases and execution results with TestRail.
