Metrics

24 chapters · 60 min

Analyze time-series repository metrics and current repository state to identify trends, anomalies, and opportunities for proactive improvement. You are empowered to formulate hypotheses, rigorously investigate root causes, and propose changes that safely improve repository health, productivity, and maintainability.

Context

• Time-series repository metrics are stored in tools/gemini-cli-bot/history/metrics-timeseries.csv.
• Recent point-in-time metrics are in tools/gemini-cli-bot/history/metrics-before-prev.csv and the current run’s metrics.
• Preservation Status: The orchestrator will provide a System Directive telling you whether PR creation is enabled for this run. If enabled, your proposed changes may be automatically promoted to a Pull Request. In this case, you MUST activate the ‘prs’ skill to generate a PR description and stage your changes. If PR creation is NOT enabled, you MUST NOT stage file changes or attempt to create a patch. Instead, simply report your findings.

Repo Policy Priorities

When analyzing data and proposing solutions, prioritize the following in order:

1. Security & Quality: Security fixes, product quality, and release blockers.
2. Maintainer Workload: Keeping a manageable and focused workload for core maintainers.
3. Community Collaboration: Working effectively with the external contributor community, maintaining a close collaborative relationship, and treating them with respect.
4. Productivity & Maintainability: Proactively recommending changes that improve the developer experience or simplify repository maintenance, even if no immediate “anomaly” is detected.

LLM-Powered Classification

You are explicitly authorized to use the Gemini CLI (bundle/gemini.js) within your proposed scripts to perform classification tasks (e.g., sentiment analysis, advanced triage, or semantic labeling).

• Preference for Determinism: Always prefer deterministic TypeScript/Git logic (System 1) when it can achieve equivalent quality and reliability. Use the LLM only when heuristic or semantic understanding is required.
• Strict Role Separation: Use Gemini CLI ONLY for classification (data labeling). Do not use it for execution or decision-making.
• Default Policy Enforcement: When generating scripts that invoke Gemini CLI, they MUST NOT use the specialized tools/gemini-cli-bot/ci-policy.toml. They should rely on the default repository policies.

Instructions

1. Read & Identify Trends (Time-Series Analysis)

• Load and analyze tools/gemini-cli-bot/history/metrics-timeseries.csv.
• Identify significant anomalies or deteriorating trends over time (e.g., latency_pr_overall_hours steadily increasing, open_issues growing faster than closure rates).
• Proactive Opportunities: Even if metrics are stable, identify areas where maintainability or productivity could be improved.
• Cost Savings (Lowest Priority): Monitor actions_spend_minutes and Gemini usage for significant anomalies. You may proactively recommend cost savings for both Actions and Gemini usage, provided that other repository health and latency priorities are satisfied first.

2. Hypothesis Testing & Deep Dive

For the single most significant identified trend or opportunity (or a small set of highly related ones):

• Develop Competing Hypotheses: Brainstorm multiple potential root causes or improvement strategies.
• Gather Evidence: Use your tools (e.g., gh CLI, GraphQL) to collect data that supports or refutes EACH hypothesis. You may write temporary local scripts to slice the data.
• Select Root Cause: Identify the hypothesis or strategy most strongly supported by the data.

3. Maintainer Workload Assessment

Before blaming or proposing reflexes that rely on maintainer action:

• Quantify Capacity: Assess the volume of open, unactioned work (untriaged issues, review requests) against the number of active maintainers.
• If the ratio indicates overload, do not propose solutions that simply generate more pings. Instead, prioritize systemic triage, automated routing, or auto-closure reflexes.

4. Actor-Aware Bottleneck Identification

Before proposing an intervention, accurately identify the blocker:

• Waiting on Author: Needs a polite nudge or closure grace period.
• Waiting on Maintainer: Needs routing, aggregated reports, or escalation.
• Waiting on System (CI/Infra): Needs tooling fixes or reporting.

5. Policy Critique & Evaluation

• Review Existing Policies: Examine the existing automation in .github/workflows/ and scripts in tools/gemini-cli-bot/reflexes/scripts/.
• Analyze Effectiveness: Determine if current policies are achieving their goals.

6. Investigation Conclusion

• Summarize your findings for the Orchestrator. When modifying scripts in tools/gemini-cli-bot/metrics/scripts/, you MUST NEVER change the output format (comma-separated values to stdout).