All authors
Claude Skills by adnanmokhtar
github.com/adnanmokhtar258 skills0 installs0 views
- Extract Conventions EmergingRound-two extraction of EMERGENT conventions — patterns that recur 5+ times across the codebase but aren't documented anywhere (error shape, pagination shape, validation pattern, transaction boundaries, async-work naming, time/money/ID handling). Used by /setup-project Phase 2.10 in REFINE mode to upgrade round-one explicit-conventions detection (file naming, suffix matrix, base classes) with the implicit "this is how we always do X" rules that the team follows by habit.Votes: 0GitHub stars: 2
- Extract Domain Entities DeeplyRound-two deep extraction of domain entities — reads ORM/model/schema definitions + migrations + repositories + tests + docs to produce a structured map (entities, fields with types/constraints, relationships, lifecycle events, invariants). Used by /setup-project Phase 2.7 in REFINE mode to upgrade round-one business-domain detection from "this is a billing app" to "billing has 7 entities with 9 invariants — here they are with file:line citations." Output is the substrate for Phase 4.6-DEEP r...Votes: 0GitHub stars: 2
- Extract Failures From HistoryRound-two: recurring failure themes from git log (revert, hotfix, incident, rollback), written as `### ` entries in ai/failures/_index.md — never per-theme files.Votes: 0GitHub stars: 2
- Extract Flows DeeplyRound-two deep extraction of business-critical and admin/internal flows — traces ≥5 representative flows end-to-end (trigger → entry → ordered steps → side effects → error paths → idempotency). Used by /setup-project Phase 2.9 in REFINE mode to upgrade round-one flow detection from "the app does checkout" to a precise step-by-step narration with file:line for every step.Votes: 0GitHub stars: 2
- Extract HotpathsRound-two identification of likely-hot endpoints / queries / jobs (heuristics — high coverage, high churn, high fan-in, monitoring mentions). For each, score N+1 risk, index coverage, cache layer, and propose a 1-line uplift. Used by /setup-project Phase 2.11 in REFINE mode to upgrade query-optimizer / parallelize-independent-ops / caching artifacts from generic prose to "here are 6 endpoints that should fix N+1 today."Votes: 0GitHub stars: 2
- Extract Project ContextGREENFIELD / Path-B project-context bootstrap — derives mission, domain, personas, KPIs, anti-goals from a user PROMPT when there's no codebase to read yet, and writes ai/project-goals.md + ai/users-and-personas.md + ai/business-domain.md. For EXISTING codebases the canonical pipeline (extract-codebase-overview → extract-business-context → Phase 4.7b) owns these targets; this skill is the greenfield path only.Votes: 0GitHub stars: 2
- Knowledge Fact ScanKnowledge claims vs lockfile and sources — wrong engine or ids, stale counts, derived copies, drifted citations.Votes: 0GitHub stars: 2
- Knowledge Ref ScanDead paths, symbols, commands, ADR ids and anchor markers in the AI knowledge layer (CLAUDE.md, ai/, anchors).Votes: 0GitHub stars: 2
- Memory Hygiene ScanLearning-sink entries — stale evidence, duplicates, contradictions, no provenance, stuck states, unreadable format.Votes: 0GitHub stars: 2
- Cutover Routing ScanStrangler cutover: façade routes, shared session, shadow isolation, one-step rollback — read-only.Votes: 0GitHub stars: 2
- Data Cutover OrchestrateV1→V2 data cutover: backfill, reconcile, load casts, ids, fenced flip, reverse path — read-only.Votes: 0GitHub stars: 2
- Extract V1 ContractExtract a V1 feature's behaviour contract — inputs, outputs, effects, rules, invariants, known bugs — every claim a pinned v1 file:line plus excerpt, prefilled from paired-inventory lane rows.Votes: 0GitHub stars: 2
- Paired InventoryPaired V1/V2 lane inventory: the same keyed rows from both trees, set-diffed into candidates — read-only.Votes: 0GitHub stars: 2
- Parity Proof ScanWas parity proven: V1-pinned tests kept, V1 branches reached, gated suites run, comparator blind spots.Votes: 0GitHub stars: 2
- Parity Test GenerateGenerate a parity test suite that runs V1 + V2 against identical inputs and asserts equivalence per the contract's tolerance taxonomy. Combines golden master, record-replay, property-based, and (when applicable) shadow / dual-write audit techniques. Outputs runnable tests and a tolerance configuration file.Votes: 0GitHub stars: 2
- Perf Uplift SurveySurvey a feature being ported from V1 to V2 for migration-time performance wins — N+1 queries, missing indexes, unbounded SELECT *, sequential awaits over independent I/O, missing caching, in-app filtering, oversized payloads. Outputs a structured candidate list with cost / saving / parity-preservation argument so each can be decided (applied / deferred / rejected) before V2 ships.Votes: 0GitHub stars: 2
- Semantics Parity ScanV1→V2 semantics parity: rounding, integer division, strings, look-alikes, order, nulls, zones — read-only.Votes: 0GitHub stars: 2
- Side Effect Parity ScanV1→V2 side-effect parity: observers, traits, dual-run duplicates, hidden consumers, crons, triggers — read-only.Votes: 0GitHub stars: 2
- Surface Parity ScanV1→V2 surface parity: routes, guards, tenant scope, enums, rules, errors, defaults — read-only.Votes: 0GitHub stars: 2
- App Size ScanFind what makes an RN/Expo, Flutter, iOS or Android release bigger than it needs to be — shrinking off, emulator ABIs, oversized or unused assets, heavy imports, debug payload, unused deps, APK uploads — each proved by a release-artifact size delta.Votes: 0GitHub stars: 2
- Device HarnessBoot a named simulator or emulator, install the app and capture evidence on both platforms — screenshots, the UI tree, cold start, a deep-link open, a process-death restore. Produces evidence, never judges it; SKIPPED when no device is available.Votes: 0GitHub stars: 2
- Mobile A11y ScanNative accessibility detector for RN/Expo, Flutter, Compose, Android Views, SwiftUI and UIKit — 8 classes, each proved by the platform's a11y checker.Votes: 0GitHub stars: 2
- Mobile Lifecycle ScanFind lifecycle defects in RN/Expo, Flutter, iOS and Android apps — state lost on process death, background and FGS limits, permission dead ends, network calls with no timeout, deep-link payload crashes, push token and cold-start tap routing, OTA runtime mismatch — each proved on a device.Votes: 0GitHub stars: 2
- Mobile Perf ScanMobile performance detector for RN/Expo, Flutter, Compose and SwiftUI — lists, identity, render waste, main-thread blocking, image decode, startup and leaks, measured in release builds.Votes: 0GitHub stars: 2
- Native Bridge AuditFind defects at the JS/Dart-to-native boundary — legacy bridge use on New-Architecture-only RN, calls the other side never answers, swallowed native errors, blocking sync methods, leaked native listeners, removed RN APIs — proved by a contract test on both sides.Votes: 0GitHub stars: 2
- Offline Sync ScanFind offline-write defects in mobile apps — RAM-only queues, replays with no idempotency key, poison heads, optimistic writes with no rollback, cross-account residue, reachability as truth — proved by a kill or a dropped response.Votes: 0GitHub stars: 2
- Platform Conventions AuditFind where an RN/Expo, Flutter, iOS or Android app breaks its platforms' conventions — edge-to-edge insets, predictive back, swipe-back, navigation stacks, safe area, per-platform drift — and close each with a /polish verb or a device-proved fix.Votes: 0GitHub stars: 2
- Store Compliance ScanProbe a mobile app's shipped config against the dated App Store and Google Play gates — privacy manifest, purpose strings, ATT, account deletion, login 4.8, target SDK, 16 KB pages, data declarations, IAP — as release blockers.Votes: 0GitHub stars: 2
- Alert AuditAlert behaviour from fire history: dead, noisy, flapping, orphaned and cause-paging alerts; NO-DATA without a backend.Votes: 0GitHub stars: 2
- Alert Rule LintPer-rule alert lint: never-fire, missing for/runbook/severity, window vs scrape, Prom 3 breakage, routing — tool-proved.Votes: 0GitHub stars: 2
- Error Tracking ScanTracker fidelity: missing release/env, no process hooks, unsymbolicated builds, uncaptured 500s — test-transport proved.Votes: 0GitHub stars: 2
- Log Signal ScanLog fidelity: invisible degrades, lost stacks, unstructured or mis-levelled logs, uncorrelated lines — captured-log proved.Votes: 0GitHub stars: 2
- Metric Hygiene ScanMetric and probe defects: unbounded labels, unmetered jobs, unit/bucket misfit, lying readiness — scrape-proved.Votes: 0GitHub stars: 2
- Slo AuditSLO attainment from live data: budget, too lax / too tight, STALE vs NO-DATA; sole writer of ai/runtime/slos.md (--no-write reads only).Votes: 0GitHub stars: 2
- Slo Rule ScanSLO definitions: no burn alert, factor vs objective/period, impossible targets, wrong SLI, bucket edge, retention — arithmetic proved.Votes: 0GitHub stars: 2
- Synthetic MonitoringBlackbox journey coverage: unprobed journeys, /health-only probes, single location, no probe-SLO, gated or dead probe alerts.Votes: 0GitHub stars: 2
- Telemetry Config ScanTelemetry config lint: collector, Loki shippers, scrape selection, dashboards.Votes: 0GitHub stars: 2
- Trace Propagation ScanTrace breaks: lost async/queue context, late SDK init, unended spans, unsafe samplers — in-memory-exporter proved.Votes: 0GitHub stars: 2
- Caching ScanFind cache defects: entries that never expire, stampedes with no single-flight, keys missing the tenant or user, writes that never invalidate, errors and nulls cached as data, HTTP cache headers, cache writes before commit, the cache as an untimed dependency. Proved by a two-tenant, write-then-read or concurrent-miss test.Votes: 0GitHub stars: 2
- Load TestDesign and run a load / stress / spike / soak campaign against an SLA on a prod-parity env, then interpret it — throughput, latency percentiles, breakpoint/knee, saturation — and return a PASS/FAIL verdict + headroom. Distinct from profile-endpoint (diagnoses ONE slow subject) and capacity-planner (analytical estimation); this is the empirical whole-system validation.Votes: 0GitHub stars: 2
- Memory Growth ScanFind server memory that grows with traffic: unbounded maps and memoisers keyed by request data, listeners and intervals never removed, request state on singletons, unreleased streams, buffers that grow while a consumer lags, units of work never cleared, heap limits against container limits. Proved by heap growth or handle count.Votes: 0GitHub stars: 2
- Memory Leak HuntHunt a memory leak by heap-diffing over time and attributing the growing allocation. Use when a process's heap/RSS grows monotonically under steady load — never bump the memory limit to hide it.Votes: 0GitHub stars: 2
- Profile EndpointLoad one endpoint at a fixed arrival rate, profile where its time goes with the runtime's sampling profiler, and capture its slow queries — runnable commands per stack. Use before optimizing.Votes: 0GitHub stars: 2
- Request Path ScanFind what slows the request path or breaks on a second replica — sequential awaits, unbounded fan-out, network N+1, blocking or CPU work on the event loop, full buffering, clients built per call, missing timeouts and load shedding, and in-process state, files, schedulers, broadcasts and jobs — each proved by a run-invariant oracle.Votes: 0GitHub stars: 2
- Web Vitals FieldMeasure Core Web Vitals in the FIELD with attribution — wire the web-vitals attribution build so a poor INP/LCP/CLS points to the exact element + sub-part. Lab tools cannot field-measure INP; this is the only path to citing the headline CWV.Votes: 0GitHub stars: 2
- Acceptance Criteria CheckAcceptance criteria: falsifiable and EARS-shaped (ears-lint), each id named by a test, no stale ids, no undefined Gherkin steps.Votes: 0GitHub stars: 2
- Assumption LedgerRank a plan's hidden assumptions by impact × uncertainty, each with its cheapest test, an expiry and a decision.Votes: 0GitHub stars: 2
- Entitlement Consistency ScanPlan entitlement detector: pricing copy vs billing vs gates, unknown gate keys, hard-coded plan names and price ids.Votes: 0GitHub stars: 2
- Evidence TraceTrace each requirement to its evidence (research, tickets, metric, commitment); names the UNSOURCED and MISLINKED set.Votes: 0GitHub stars: 2
- Feature Surface ScanHalf-built feature detector: unserved client calls, uncalled routes, dead links, shipped placeholders, dead buttons.Votes: 0GitHub stars: 2