Agent
Backend analysis, architecture review, and implementation guidance.
Agent
Backend analysis, architecture review, and implementation guidance.
Agent
Coordinates multi-agent AI reasoning tasks.
Agent
Requirements analysis, tracing, and validation.
Agent
Test generation, validation, and quality assurance.
Instructions file GitHub Copilot
Instructions for rijuvashisht/Magneto, covering copilot instructions — magneto ai, rules, available agents and mcp tools.
Skill Claude CodeCodex
Role: Audit Log Reader Inputs: filters (sprint, skill, time range) Outputs: human-readable log or JSON Audit actions: audit.read, audit.verified.
Skill Claude CodeCodex
Role: Review Pipeline Inputs: plan.md Outputs: plan-reviews.md with all reviews + test-plan.md Audit actions: autoplan.started, autoplan.completed.
Skill Claude CodeCodex
Role: Performance Engineer Inputs: URL(s) to benchmark Outputs: benchmark report in .magneto/benchmarks/ Audit actions: benchmark.baseline, benchmark.compared.
Skill Claude CodeCodex
Role: SRE Inputs: deployed URL Outputs: continuous monitoring report; alerts on anomalies Audit actions: canary.started, canary.anomalydetected, canary.cleared.
Skill Claude CodeCodex
Role: Safety Guardrail Inputs: none Outputs: updates .magneto/safety.json Audit actions: careful.activated, careful.warned, careful.overridden.
Skill Claude CodeCodex
Role: Cross-Model Reviewer Inputs: current diff or target code Outputs: review-report.md with multi-model analysis Audit actions: codex.started, codex.convergencescored, codex.completed.
Skill Claude CodeCodex
Role: Drift Detector Inputs: spec docs + current codebase Outputs: drift report Audit actions: drift.scanned, drift.reported, drift.reconciled.
Skill Claude CodeCodex
Role: Edit Lock Inputs: one or more path globs Outputs: updates .magneto/safety.json Audit actions: freeze.activated, freeze.blockedwrite.
Skill Claude CodeCodex
Role: Full Safety Inputs: one or more path globs Outputs: updates .magneto/safety.json (activates both careful + freeze) Audit actions: guard.activated.
Skill Claude CodeCodex
Role: Debugger Inputs: bug description or error trace Outputs: review-report.md with root-cause analysis Audit actions: investigate.started, investigate.hypothesistested, investigate.resolved.
Skill Claude CodeCodex
Role: Release Engineer Inputs: open PR (from /ship) Outputs: merged PR + deploy verification Audit actions: land-and-deploy.merged, land-and-deploy.deployed, land-and-deploy.verified.
Skill Claude CodeCodex
Role: Memory Manager Inputs: optional query, tags, or new learning Outputs: .magneto/memory/ updates Audit actions: learn.recorded, learn.pruned, learn.exported.
Skill Claude CodeCodex
Role: CEO / Founder Inputs: plan.md Outputs: appends to plan-reviews.md Audit actions: plan-ceo-review.scored, plan-ceo-review.recommended.
Skill Claude CodeCodex
Role: Senior Designer Inputs: plan.md Outputs: appends to plan-reviews.md Audit actions: plan-design-review.scored, plan-design-review.polished.
Skill Claude CodeCodex
Role: Eng Manager Inputs: plan.md Outputs: appends to plan-reviews.md, creates test-plan.md Audit actions: plan-eng-review.scored, plan-eng-review.testmatrixdrafted.
Skill Claude CodeCodex
Role: Eng Manager + CEO Inputs: design.md (from /think) Outputs: plan.md Next skill: /review (or /autoplan for full review pipeline) Audit actions: plan.started, plan.scoped, plan.drafted.
Skill Claude CodeCodex
Role: QA Reporter Inputs: test-plan.md + target URL Outputs: qa-report.md only — no code changes Audit actions: qa-only.started, qa-only.bugfound.
Skill Claude CodeCodex
Role: QA Lead Inputs: test-plan.md (from /plan-eng-review) + target URL Outputs: qa-report.md, regression tests in tests/ Audit actions: qa.started, qa.bugfound, qa.bugfixed, qa.testsgenerated.
Skill Claude CodeCodex
Role: Eng Manager Inputs: audit log + sprint artifacts Outputs: retro.md Audit actions: retro.started, retro.drafted.