SylphxAI
OrganizationReusable, organization-neutral Agent Skills for product, engineering, operations, design, and research work.
Categories
Indexed Skills (122)
accessibility-conformance-program-review
Accessibility conformance program: AT coverage, release evidence, VPAT/ACR, remediation ownership.
account-recovery-review
End-to-end account recovery: lost credentials/MFA, locks, compromise, ownership proof, post-recovery hardening.
ad-monetization-review
In-product ad system: formats, mediation, rewards, consent, frequency caps, fraud, IAP/IAA tradeoffs.
agent-context-engineering
Agent context and memory contract: ranking, compaction, invalidation, provenance, freshness, privacy.
agent-planning-system-review
Agent planning and control machinery → Planning and Control Contract.
agent-system-improvement
Recurring agent-system failures: Learning Change Record, controlled intervention, replay/holdout, promotion.
ai-product-risk-review
AI product behavior/launch contract: job, autonomy bounds, data use, failure/abuse, disclosure, fallback, cost.
app-design-blueprint
Whole-app design blueprint for new apps or full redesigns: jobs, IA, activation, retention, multi-surface.
board-metrics-operating-review
Board/executive metrics operating model: definitions, lineage, decision use, reporting cadence, trust.
causal-inference-analysis
Causal effects: identification strategy, assumptions, estimators, robustness, claim bounds.
convergent-review
Converge divergent analysis into one decision record with explicit residuals and commitments.
critical-analysis
Adversarial critical analysis of a claim, plan, or design with kill criteria and residuals.
customer-data-migration-review
Customer data migration: mapping, dual-write/read, cutover, rollback, integrity proof.
customer-success-operations-review
Customer success operating model: segments, plays, health scores, expansion, handoffs.
customer-support-case-resolution
Resolve one support case to a durable outcome with evidence and customer-visible close.
customer-support-operations
Support operations: queues, SLAs, escalation paths, tooling, quality sampling.
daily-reward-and-streak-review
Daily rewards and streaks: incentives, fairness, abuse, retention coupling.
data-quality-observability-review
Data quality observability: contracts, monitors, incident paths, ownership.
data-rights-operations-review
Data-rights operations: DSAR, retention, deletion, fulfillment proof.
decision-optimization-modeling
Decision optimization model: objective, constraints, solver choice, sensitivity.
dependency-version-selection
Select dependency and runtime versions from live authoritative release sources.
design-space-exploration
Map a design space: alternatives, dimensions, kill criteria, ranked options.
design-system-extractor
Extract design system tokens/components/rules from live product UI.
developer-product-experience-review
Developer product experience: onboarding, APIs, docs, DX friction, time-to-first-success.
developer-quota-credit-abuse-review
Developer quota and credit abuse: metering, limits, fraud patterns, enforcement.
enact-work-coordination
Live Enact Work with authenticated tools: propose, claim, run, evidence, effects, defer/wake.
engineering-standard
Engineering method floor: design, verification, change safety, reproduction-driven fixes.
enterprise-access-governance-review
Enterprise access governance: roles, joiner-mover-leaver, access reviews, least privilege.
enterprise-contract-operations-review
Enterprise contract operations: obligations, renewals, compliance hooks, negotiation artifacts.
evidence-and-claims-standard
Claim grading: evidence class, provenance, strength, promotion boundaries.
evidence-synthesis
Synthesize multi-source evidence into one bounded synthesis with residuals.
game-design-blueprint
Whole-game design/GDD: promise, core loops, progression, economy, FTE, live ops.
game-economy-review
Game economy: sources/sinks, balance, progression coupling, abuse vectors.
game-soft-launch-review
Game soft-launch readiness: markets, KPIs, kill/scale criteria, live-ops plan.
high-signal-communication
Compress or restructure an agent communication into a concise, decision-complete, high-signal artifact. Use when the user explicitly asks to shorten, de-duplicate, hand off, summarize progress, reduce context cost, or repair a verbose/slow-to-understand reply, report, commit, PR, ADR, spec, incident note, or agent record. Do not inject by default for every ordinary reply or document.
architecture-convergence
Substantial architecture rewrite onto shared invariants with migration path and acceptance evidence.
frontend-performance-remediation
Frontend performance remediation: measurement, budgets, critical path, regression control.
autonomous-execution
Drive one accepted multi-phase objective through research, execute, verify, repair to delivery-terminal evidence.
composable-product-portfolio
Multi-product portfolio seams: what to share vs isolate, composition rules across products.
continuous-product-quality
High-leverage product betterment: coverage card, admit all B, execute, verify; continue until no unblocked high-EV work remains.
analyze-causal-inference
Estimate causal effects: identification, estimators, robustness bounds.
analyze-critically
Adversarially critique a claim/plan/design with kill criteria.
analyze-system-dynamics
Map stubborn loops: stocks, delays, incentives, policy resistance.
author-skill
Create or revise an Agent Skill package: job, description, method, references.
curate-skill-repository
Curate skill portfolio: merge/split, collisions, retirement.
forecast-with-calibration
Produce calibrated probabilistic forecasts with base rates and ranges.
maintain-product
Cut live/regression harm with root-cause fix and oracle.
optimize-decision-model
Model a decision for optimization: objective, constraints, sensitivity.
produce-game-2d-sprites
Produce engine-ready 2D game sprites/sheets with QC.
record-structured-deliberation
Record a structured deliberation: options, criteria, decision, actions.
execute-hard-cutover
Hard-cut predecessor to destination sole writer; retire dual paths.
operate-customer-support
Operate customer support: triage, resolution, escalation, quality loops.
optimize-store-listing
Optimize store listing: metadata, creatives, conversion, compliance.
price-saas-subscription
Price a SaaS subscription: packaging, meters, fences, migration.
adopt-repo-standards
Adopt engineering baseline into a repo: constitution, manifest, fences.
compose-product-portfolio
Compose multi-product portfolio: platform vs products, connectors, bundles.
compose-product-program
Compose multi-work product program: phases, dependencies, kill/scale.
craft-human-agent-language
Write or review any document or prompt that humans and agents both use (README, ADR, update, instruction): frame purpose and audience, structure before prose, plain language, model-agnostic language choice, output contract, and a re-verification gate. Use when asked to write/translate/review a doc or prompt for humans and agents, or to choose the most efficient language and format for a model you are switching.
craft-product-interface
Craft one product UI flow with states, a11y, verification.
decide-architecture-shape
Decide one architecture shape from options with tradeoffs.
design-product
Design a product blueprint (app, game, SaaS, etc.); shape depth in references.
design-skill-evals
Design falsifiable skill evals: routing cases, oracles, failures.
distill-source-to-skill
Distill docs/repos/notes into one installable Skill package.
weekly-capacity-review
Run the weekly platform-team capacity review: collect the three inputs, compute per-engineer utilization flags, triage the backlog into committed/candidate/parked, and record decisions and follow-ups. Use when asked to prepare, run, record, or follow up the weekly capacity review, update the capacity sheet, or triage the queue for next week. Not for incident response or one-off backlog grooming.
my-distilled-skill
Run the weekly platform capacity review: collect capacity/backlog inputs by Thursday 12:00, compute per-engineer utilization flags (above 90% or below 40% for three consecutive weeks), rank the backlog into committed/candidate/parked, and record decisions and follow-ups. Use when asked to run, prepare, or audit a weekly capacity review, capacity utilization analysis, or backlog prioritization. Not for incident response (separate runbook) or sprint planning.
edit-preserving-voice
Edit text preserving voice, constraints, and factual claims.
engineer-agent-context
Engineer agent context packs: load rules, budgets, refresh.
engineer-testable-requirements
Turn requirements into testable acceptance criteria with oracles.
model-security-threats
Model security threats for a trust boundary before implementation.
research-public-web
Research the public web without paid SERP keys; cite sources and record adapter gaps.
skills
Install, update, or verify the complete SylphxAI/skills managed environment for the current Codex, Claude Code, or Grok runtime, including AutoSync and canonical Sylphx Enact SaaS enrollment. Use when a user gives this repository and says "install this", asks to update Sylphx Skills, or requests installation readback. Read the repository's INSTALL.md before choosing any generic Skill installer; folder copies alone are incomplete, and other runtimes stay untouched unless the user explicitly includes them.
agent-first-development-standard
No-human multi-agent delivery packets: change attribution, collision scopes, machine-verifiable acceptance.
agent-native-standard
Agent-operable products: queryable typed state, schema-derived tools, recoverable workflows, run traces.
autonomous-execution-standard
Autonomous scheduling: critical-path focus, work-conserving progress, complexity-gated resource-safe delegation, replanning, and honest evidence layers. Use when a non-trivial task should continue through multiple independent work lanes without repeated user direction. Do not use for a short direct answer, a few file reads, one command, or a tightly coupled local edit.
ci-admission-standard
Protected-branch CI admission: required checks, risk-matched gates, flake policy, preview gates.
ci-runner-capacity-standard
CI runner pools: fairness, isolation, scaling, queue pressure under load.
commercial-decision-standard
Durable commercial decisions: pricing, packaging, entitlements, licensing, route-to-market ADRs.
decision-quality-standard
Decision-process quality: framing, alternatives, information, tradeoffs, commitment, learning.
delivery-standard
Delivery-terminal proof: land, artifacts, publish, deploy, rollback, live readback.
documentation-standard
Documentation create/generate/retire: minimal sufficient, freshness, low prose drift.
enterprise-control-plane-standard
Org-wide multi-repo control plane: rulesets, reconcilers, audits, status projections.
enterprise-profile-standard
Versioned enterprise profiles: selectors, defaults, exceptions, replacement, migration.
frontier-verification-standard
Evaluation gates, replay oracles, and canary backstops for nondeterministic systems. Use when deterministic unit and integration tests cannot fully verify AI, probabilistic, distributed, or time-dependent behavior. Use CI Admission for flaky-check attribution or quarantine; do not add frontier mechanisms to an ordinary deterministic code path.
instruction-evolution-standard
Evolution of Skills, runtime constitutions, required profiles, and architecture generations with one writable owner, migration, reconciliation, and retirement. Use for durable instruction or profile changes, not live Work Graph operations.
accessibility-conformance
Assess whether disabled users can complete the selected tasks, and what the current results support saying about conformance. Use when the user asks to review accessibility, audit a conformance claim, or check a VPAT or accessibility statement. Do not use to design the system under review or to fix accessibility defects.
app-product
Design an app product whose promise is a recurring job — thesis, journeys, system shape, platform behavior, and validation. Use when the user asks to design an app, define an app concept, or shape a mobile, desktop, or web app. Do not use for a game, a SaaS web platform, a single UI flow, or a pricing catalog.
bound-request-scope
Bound this request: objective, in/out, non-goals, cut lines.
build-distribution-readiness
Prepare and verify a signed product package for its distribution channels — install, update, uninstall, and live readback. Use when shipping a binary, installer, package, or store build. Do not use to write store listing copy or screenshots.
build-payment-readiness
Design and verify production payments — checkout, provider events, ledger, entitlements, refunds, disputes, settlement, and finance close. Use when the user mentions Stripe, app-store billing, webhooks, invoices, chargebacks, or taking money in production. Do not use for seller marketplace payouts, first-time packaging, or a live price increase.
build-product
Build a missing product capability end to end on a real user path. Use when an accepted direction needs a usable slice across code, interface, data, and delivery. Use the product's current identity, storage, job, and event providers; do not invent a second runtime.
craft-telegram-bot-surface
Design or implement a Telegram bot on the current Bot API. Use when the user mentions Telegram, Bot API, slash commands, inline keyboards, callback queries, Web Apps, sendRichMessage, or rich messages. Do not use parse_mode HTML/MarkdownV2 as the product text path. Do not use for Slack, Discord, or generic product UI.
data-rights-request
Fulfill one access, export, correction, deletion, restriction, or appeal request with proof and reconciliation. Use when a user asks to fulfill a data-rights request, a DSAR, a GDPR or CCPA request, or a specific access/delete ticket. Do not use to design the privacy lifecycle, and not for legal advice.
design-offline-sync
Design an offline sync and conflict protocol that preserves user intent across retries, multi-device edits, deletion, and partial failure. Use when offline is a declared product capability. Do not use for ordinary online CRUD or for a live incident.
design-price-increase-retention
Design a live subscription price increase so contract, notice, consent, billing channel, and no-action default stay truthful. Use when raising prices on existing contracts. Do not use for first-time packaging or for implementing checkout.
design-product-experiment
Turn a product uncertainty into a pre-registered experiment protocol with an estimand, assignment, guardrails, and a decision rule. Use when the result must be able to change a decision. A staged rollout is not automatically an experiment. Do not use to estimate an already-identified observational effect or to run a demo without assignment.
drive-to-delivery
Drive one already accepted engineering or product objective to its requested terminal. Use when completion matters more than a plan. Do not use to choose what to do next from a backlog.
finish-product
Finish an already integrated product to a declared target. Use when the promise is clear and the job is deep resolution of copy, states, feel, and consistency. Do not use to add a capability or to repair a live defect.
game-product
Design a game product whose promise is play, mastery, fantasy, or progression — loops, systems, feel, and validation. Use when the user asks to design a game, define a game concept, or shape game loops or systems. Do not use for an app, a SaaS web platform, a single UI flow, or a game economy review.
handoff-work
Safely persist in-progress state and write an actionable resumption brief for another machine, session, or developer. Use when handing off work, switching computers or workspaces, pausing mid-task, saving work-in-progress across environments, or transferring an unfinished objective. Do not use for routine status updates without environment transfer (use write-high-signal-update), fully completed deliverables ready to merge (use drive-to-delivery), or production cutovers (use execute-hard-cutover).
launch-readiness
Assess one release candidate for a go/no-go, hold, or risk decision on the exact build. Use when the user asks whether a release, launch, or rollout is ready, or to review a launch plan. Do not use to plan or execute the launch.
Showing top 100 of 122 skills by quality score. Browse the full catalog at /browse.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.