Draft anonymous one-page company teasers for sell-side M&A processes. Creates a compelling summary without revealing the company's identity, designed to gauge buyer interest before NDA execution. Triggers on "teaser", "blind teaser", "anonymous profile", "one-pager for process", or "draft teaser for sell-side".

Investigates distributed application performance using PostHog APM (OpenTelemetry span) data via MCP. Use when the user asks about service traces, slow HTTP/database spans, error spans, error-rate trends or spikes, latency distributions, trace IDs, or span attributes — not AI observability traces or product logs. Uses posthog:query-apm-spans, posthog:apm-trace-get, posthog:apm-spans-sparkline, posthog:apm-services-list, posthog:apm-attributes-list, and posthog:apm-attribute-values-list.

Clarify how to visualize change over a time range before building a trend. Use whenever the user asks how much something changed, grew, dropped, improved, or regressed between two points or periods — "how much did X change from A to B", "before vs after", "start vs end", "week over week", "compare this month to last", "change over time" — or mentions a "slope chart" / "slopegraph". Two readings of "change" need different charts: the whole trend (a line, every interval) versus just the two endpoints (a slope, start vs end). Ask which they want, then render it. Not for choosing a saved insight ChartDisplayType in the insight editor.

Explore PostHog MCP intent clusters — agent goals grouped by semantic similarity, with each cluster's tool distribution and error rates, plus the tool-centric pivot (capture rate per intent, discovery rate against the advertised catalog, description fit, tool overlaps). Use when the user asks "what are agents trying to do with the MCP?", "group the intents", "which goals fail most?", "what does each cluster route to?", "when agents have this intent do they find my tool?", "which tools get mixed up?", wants to recompute the clustering, or pastes an MCP analytics intent-clustering URL.

Guides experiment state transitions: launching, pausing, resuming, freezing/unfreezing exposure, ending, shipping variants, archiving, resetting, duplicating, and copying to another project. Covers preconditions, implications for variant assignment and analysis, and the decision framework for when to use each action. TRIGGER when: user asks to launch, pause, resume, end, ship, archive, reset, duplicate, or copy an experiment to another project, or to freeze/unfreeze exposure (stop enrolling new users while metrics keep flowing, or reopen enrollment). DO NOT TRIGGER when: user is creating an experiment (use creating-experiments), configuring rollout (use configuring-experiment-rollout), or setting up metrics (use configuring-experiment-analytics).

Diagnose PostHog Surveys configuration and responses across all five SDKs (web/posthog-js, iOS, Android, Flutter, React Native). Use whenever a Surveys support ticket is pasted ("survey not showing", "fewer responses than expected", "responses disappeared", "responses are incomplete", "only the first question was answered", "the user says they didn't mean to submit", "survey shows on wrong platform"), or when diagnosing why a survey does or doesn't display. Covers the eligibility pipeline, how a response actually gets stored (partial responses, branching, optional questions, auto-submit), cross-SDK feature parity, the known-cause catalog, read-only diagnostic queries, staff access, and the customer-reply style guide.

Investigate LLM spend in PostHog — total cost over time, cost by model, provider, user, trace, or custom dimension, token and cache-hit economics, and cost regressions. Use when the user asks "how much are we spending on LLMs?", "which model / user / feature is most expensive?", "why did cost spike?", wants to build a cost dashboard or alert, or pastes a trace URL and asks about its cost.

Find where an AI/LLM application is failing in production and surface the failure patterns, working from real traces. Use when someone wants to understand what's going wrong with an AI feature, find and categorize failure modes, triage errors, or investigate quality issues (wrong answers, ignored instructions, hallucinations, tool misuse) — "what's failing in my agent", "surface error patterns", "why are the responses bad", "find the common failure modes", "what should I fix next". Covers scoping to one use case, finding failing traces by whichever signal fits the context (code errors, metric outliers, trace-type slices, manual review, existing-eval spikes, clustering), and reading them into a ranked failure taxonomy.

Author continuously-running online evaluations in PostHog AI observability, grounded in real failure modes you've identified. Use when the user wants evaluations that automatically score new generations or whole traces going forward — "create an eval to catch X", "continuously check that responses do Y", "turn these failures into evals". Covers letting the explored data decide how many evals to create, proposing that set for the user to pick, choosing the target and eval type (hog / llm_judge / sentiment), configuring a provider and model for an llm_judge eval (a provider key gates enabling, not creation), scoping which generations trigger it via conditions, creating disabled, verifying scope, and enabling. Proposes a sentiment eval when no failure mode is worth catching. Finding and ranking the failure modes worth evaluating is its own job — use exploring-ai-failures first. To debug or manage evaluations that already exist, use exploring-llm-evaluations.

Helps PostHog engineers and agents pick the right chart when a request is about visualizing change over time, before/after comparisons, or period-over-period movement while building UI with @posthog/quill-charts (dashboards, reports, the mcp_analytics frontend, custom visualizations). Use when a user asks how something "changed", "moved", "grew", "dropped", "improved", or "regressed" between two points or periods, or mentions "before/after", "period over period", "week over week", "delta", "slope graph", "slopegraph", or comparing many series across two snapshots. Surfaces both the `SlopeChart` quill component and the native `ChartDisplayType.SlopeGraph` insight display as options.

Get unstuck with Watt without leaving Claude — ask what you can do and how, figure out why something isn't working, see what's changed and whether you're on the latest, or tell the team something — report a bug, request a signal or feature, reach a human, and track what you've filed. It answers your question first and gets you to a result fast — pointing you to /watt:explore, /watt:audience, or /watt:configure when that's the quicker path. Use when you type /watt:help, or say "what can Watt do", "how do I build an audience", "why is X failing", "Claude is blocking my export", "something's broken", "I need a signal for Y", "I need a human", "what's new", or "what have I filed".

Build a person audience from an owned list — customers, leads, accounts — resolving it to Watt entity IDs and emitting a roster, or learning what defines it and handing back a tunable signal pool. The list-anchored way into a build, behind /watt:audience. Four plays — resolve-only, expand (households via addresses), lookalike (the signals that define the list, to tune), overlay (score and rank against signals); group and traverse aren't built. Produces a roster of entity IDs or a signal pool — never an export, never contact data. Not a user command — /watt:audience is the front door. Use when a list arrives to build from — "match my customer list", "find more like my customers", "rank my list by intent", "score the people from my last run". A read intent is audience-analyze-list.

Build a person audience from a plain-English description — discover the signals behind the brief, score them, and compose, partition, or cross them into a measured result. The discovery-driven way into a build, behind /watt:audience; anchored on a description (no owned list). The objective picks how it lands — a size band, the widest credible reach, the highest-precision few, grouping into the best-concentrated cells, or crossing the employment graph to their employers (or employees). Produces a signal stack + measured reach, or a roster — never an export, never contact data. Not a user command. Use when a build starts from a description — "build me an audience of about 1–5M weekend hikers", "the highest-intent people in-market for solar", "where do these people cluster", "a lead list of in-market companies".

Scan or clean prose for AI-generated residue, including unsupported connective tissue, epistemic inflation, over-completed arguments, performed voice, formulaic structure, empty figurative language, stock phrasing, corporate abstraction, hedging, and false enthusiasm. Use for AI-smell cleanup, humanizing model-assisted prose, source-faithful drafting validation, writing polish, or pre-publication checks; enforce silently when composed into drafting or line editing.

Write, review, or organize documentation following the Diátaxis framework. Use when writing docs, reviewing docs, creating tutorials, how-to guides, reference material, or explanatory content.

Search Google News, Google Shopping, Google Images, and Google Lens using Serper APIs via x402. USE FOR: - Finding recent news articles on a topic - Current events and breaking news - Product searches and price comparisons - Shopping research and product discovery - Industry news monitoring - Image search and finding photos/headshots of people - Reverse image search (Google Lens) TRIGGERS: - "news about", "latest news", "recent articles" - "current events", "breaking", "headlines" - "shopping", "buy", "price", "product search" - "compare prices", "where to buy", "deals on" - "image of", "photo of", "headshot", "reverse image search" Use agentcash.fetch for Serper endpoints. Most endpoints are $0.04 per call; Lens is $0.20.

Make AI phone calls, buy phone numbers, and get call transcripts via x402. USE FOR: - Making AI-powered outbound phone calls - Scheduling calls and getting transcripts - Buying and managing phone numbers - Leaving voicemails - Transferring calls to humans - Checking if a number has iMessage/FaceTime TRIGGERS: - "call", "phone call", "make a call", "dial" - "buy number", "phone number", "get a number" - "voicemail", "leave a message" - "transcript", "call recording", "call summary" - "AI call", "automated call", "bland" - "imessage", "facetime", "check number", "lookup number" ALWAYS use agentcash.fetch for stablephone.dev endpoints. IMPORTANT: Use exact endpoint paths from the Quick Reference table below.

Explain how the code on the current branch works, as an HTML page opened in the browser, written so someone can understand the change and review it. Use when asked to explain a branch, explain these changes, walk me through this code, help me review this, or explain what an agent just did. Most useful when someone other than the reader wrote the code and it has to be understood before merging.

Generate or update README.md documentation for Helm charts in this project. Extracts all values from values.yaml, describes deployed resources, and produces a complete README with architecture, installation, configuration table, and upgrade sections. Use when: documenting a chart, generating a README, updating docs after adding features or values. Trigger phrases: "document chart", "generate README", "update docs for chart", "generate documentation", "write README for chart".

Reference for paired-reversal handling on GL batch tapes: an RV row whose ref-trace points at an earlier row cancels BOTH legs (the RV and the row it references) — but ONLY when the two value-dates fall within the shop's reversal settlement window; an RV that references a too-old posting is NOT a cancellation and posts one-sided. Useful when the posting tape carries a reference-trace column alongside reversal codes and the running balance is off by twice the reversal magnitude, or by a whole leg on stale references.