manolii-framework

Navigate external websites via the best available browser automation path

Monitor a deployment for errors after shipping

Toggle session-scoped slow-mode (smaller diffs, more confirmation)

Capture current work state for session resumption

Read .ai/compact-metrics.jsonl and generate a health report on how the smart compaction system is performing

Broad agent: queries graphify for downstream caller counts and god-node status on changed public symbols. Flags breaking change risk.

CI failure investigator — reads PR diff and CI log excerpts, diagnoses root cause, proposes a scoped fix. Read-only, propose-only. Never auto-applies fixes.

Delegates to OpenAI Codex in adversarial mode for cross-provider review. Challenges implementation choices.

Deep knowledge synthesis across project memory for decisions that need the full picture of a topic or area

Deep analysis specialist — architecture assessments, technical reports, codebase walkthroughs, cross-repo impact analysis.

Aggregates telemetry into a weekly analytics report: tool usage, error rates, compaction patterns, and day-of-week activity.

Run the Model Change Protocol — 15-item assessment whenever you add, remove, or re-route any AI model in your project. Produces a decision log and assessment artefact. Mandatory before any commit that touches .claude/model-routing.json or agent/skill model declarations.

Real-browser testing methodology for catching bugs that unit tests and CI miss. Uses Playwright for automated browser interaction with visual verification.

Structured invariant verification for active multi-phase projects. Runs four independent checks: phase gates, datastore write health, required files, checkpoint freshness. Always advisory.

Checks config.yaml, fly.toml, and routing JSON diffs for missing sibling parameters — when one entry in a group gets a param, all siblings must too.

9 hooks across 7 events