Skip to content

SKILL.md

Latest commit

 

History

History
303 lines (207 loc) · 12.3 KB

SKILL.md

File metadata and controls

303 lines (207 loc) · 12.3 KB
Raw
name pp-posthog
description Every PostHog resource in one CLI — with offline search, agent-native output, and cross-resource analytics no... Trigger phrases: `check my PostHog feature flags`, `query PostHog events`, `show experiment results in PostHog`, `what errors are spiking in PostHog`, `LLM costs in PostHog`, `is it safe to ramp this flag`, `use posthog`.
author riteshtiwari
license Apache-2.0
argument-hint <command> [args] | install cli|mcp
allowed-tools Read Bash
metadata
openclaw
requires
bins
posthog-pp-cli

PostHog — Printing Press CLI

Prerequisites: Install the CLI

This skill drives the posthog-pp-cli binary. You must verify the CLI is installed before invoking any command from this skill. If it is missing, install it first:

  1. Install via the Printing Press installer: 
    npx -y @mvanhorn/printing-press-library install posthog --cli-only
  2. Verify: posthog-pp-cli --version
  3. Ensure $GOPATH/bin (or $HOME/go/bin) is on $PATH.

If the npx install fails (no Node, offline, etc.), fall back to a direct Go install (requires Go 1.26.3 or newer):

go install github.com/mvanhorn/printing-press-library/library/developer-tools/posthog/cmd/posthog-pp-cli@latest

If --version reports "command not found" after install, the install step did not put the binary on $PATH. Do not proceed with skill commands until verification succeeds.

When to Use This CLI

Use posthog-pp-cli when an agent task requires querying, diffing, or combining PostHog data across more than one resource type — flag state vs error events, LLM costs vs variant performance, experiment exposure vs significance. It is faster than the web UI for bulk reads and produces structured JSON that agents can consume directly without parsing HTML.

Unique Capabilities

These capabilities aren't available in any other tool for this API.

Flag safety

  • flags blast-radius — Find every insight, dashboard, experiment, and survey that references a flag before you archive or rename it.

    Use before archiving or renaming a flag to prevent breaking dashboards and experiments silently.

    posthog-pp-cli flags blast-radius --key my-checkout-v2 --json

  • flags rollout-health — Go/no-go confidence for a flag ramp — error rate and key metric movement correlated with flag exposure.

    Use before ramping a flag to 100% to catch regressions that aren't visible in overall metrics.

    posthog-pp-cli flags rollout-health --key new-checkout --window 24h --agent

  • flags stale — List flags that haven't been evaluated in N days — cleanup candidates before they accumulate.

    Use in quarterly flag cleanup sprints to identify dead code paths safely.

    posthog-pp-cli flags stale --days 30 --json

LLM observability

  • llm cost-attribution — Break down LLM spend by feature flag variant — see whether the expensive model variant pays for itself.

    Use when evaluating whether to promote a more expensive LLM variant based on actual cost-per-conversion.

    posthog-pp-cli llm cost-attribution --flag model-tier --agent

Local state that compounds

  • persons at-risk — Surface which users in a cohort are going quiet and recently hit errors — before they churn.

    Use in weekly retention reviews to prioritize proactive outreach before churn events.

    posthog-pp-cli persons at-risk --cohort paying-users --silent-days 14 --json

  • events property-drift — Catch tracking regressions — properties that silently disappeared from an event between two time windows.

    Use after a deploy to catch silent schema changes that corrupt ongoing experiments and dashboards.

    posthog-pp-cli events property-drift checkout_completed --agent

  • experiments pre-check — Know today whether an experiment will reach significance this sprint, or needs traffic adjustment now.

    Use at the start of each sprint to surface experiments that need traffic changes before they run out of time.

    posthog-pp-cli experiments pre-check --agent

  • dashboard health — Find broken dashboards before a stakeholder meeting does — stale data, deleted cohorts, archived flags.

    Use before weekly business reviews to catch broken insight tiles that would embarrass the data team.

    posthog-pp-cli dashboard health --stale-days 7 --agent

Command Reference

projects — Manage projects

public-hog-function-templates — Manage public hog function templates

  • posthog-pp-cli public-hog-function-templates — List

user-home-settings — Manage user home settings

  • posthog-pp-cli user-home-settings partial-update — Update the authenticated user's pinned sidebar tabs and/or homepage for the current team. Pass @me as the UUID....
  • posthog-pp-cli user-home-settings retrieve — Get the authenticated user's pinned sidebar tabs and configured homepage for the current team. Pass @me as the UUID.

users — Manage users

  • posthog-pp-cli users cancel-email-change-request-partial-update — Cancel email change request partial update
  • posthog-pp-cli users destroy — Destroy
  • posthog-pp-cli users list — List
  • posthog-pp-cli users partial-update — Update one or more of the authenticated user's profile fields or settings.
  • posthog-pp-cli users request-email-verification-create — Request email verification create
  • posthog-pp-cli users retrieve — Retrieve a user's profile and settings. Pass @me as the UUID to fetch the authenticated user; non-staff callers...
  • posthog-pp-cli users update — Replace the authenticated user's profile and settings. Pass @me as the UUID to update the authenticated user....
  • posthog-pp-cli users verify-email-create — Verify email create

Finding the right command

When you know what you want to do but not which command does it, ask the CLI directly:

posthog-pp-cli which "<capability in your own words>"

which resolves a natural-language capability query to the best matching command from this CLI's curated feature index. Exit code 0 means at least one match; exit code 2 means no confident match — fall back to --help or use a narrower query.

Recipes

Find flags safe to archive

posthog-pp-cli flags stale --days 60 --json | jq '.[].key'

List flag keys with no evaluation events in 60 days — safe cleanup candidates

Pre-ramp safety check

posthog-pp-cli flags rollout-health --key new-checkout --window 48h --agent --select flag_key,error_rate_delta,metric_delta

Check error rate and purchase metric delta for flag-exposed users before ramping to 100%

Weekly experiment briefing

posthog-pp-cli experiments pre-check --json | jq '.[] | {key, winner, significance, days_remaining}'

Cross-experiment pre-check summary for Monday standup — surface experiments needing traffic adjustment

LLM cost by flag variant

posthog-pp-cli llm cost-attribution --flag model-tier --days 30 --agent --select variant,total_cost_usd,avg_cost_per_call

Compare LLM spend across A/B variants to decide if GPT-4o outperforms GPT-4o-mini in your context

Dashboard audit before all-hands

posthog-pp-cli dashboard health --stale-days 7 --json | jq '.[] | select(.issues | length > 0)'

Surface dashboards with broken filters or stale data before a stakeholder presentation

Auth Setup

Uses your PostHog personal API key (phx_...). Set POSTHOG_API_KEY or run posthog-pp-cli auth set-token. Supports both US (app.posthog.com) and EU (eu.posthog.com) instances via POSTHOG_HOST.

Run posthog-pp-cli doctor to verify setup.

Agent Mode

Add --agent to any command. Expands to: --json --compact --no-input --no-color --yes.

  • Pipeable — JSON on stdout, errors on stderr

  • Filterable — --select keeps a subset of fields. Dotted paths descend into nested structures; arrays traverse element-wise. Critical for keeping context small on verbose APIs:

    posthog-pp-cli flags stale --agent --select key,days_stale

  • Previewable — --dry-run shows the request without sending

  • Offline-friendly — sync/search commands can use the local SQLite store when available

  • Non-interactive — never prompts, every input is a flag

  • Explicit retries — use --idempotent only when an already-existing create should count as success, and --ignore-missing only when a missing delete target should count as success

Response envelope

Commands that read from the local store or the API wrap output in a provenance envelope:

{
  "meta": {"source": "live" | "local", "synced_at": "...", "reason": "..."},
  "results": <data>
}

Parse .results for data and .meta.source to know whether it's live or local. A human-readable N results (live) summary is printed to stderr only when stdout is a terminal — piped/agent consumers get pure JSON on stdout.

Agent Feedback

When you (or the agent) notice something off about this CLI, record it:

posthog-pp-cli feedback "the --since flag is inclusive but docs say exclusive"
posthog-pp-cli feedback --stdin < notes.txt
posthog-pp-cli feedback list --json --limit 10

Entries are stored locally at ~/.posthog-pp-cli/feedback.jsonl. They are never POSTed unless POSTHOG_FEEDBACK_ENDPOINT is set AND either --send is passed or POSTHOG_FEEDBACK_AUTO_SEND=true. Default behavior is local-only.

Write what surprised you, not a bug report. Short, specific, one line: that is the part that compounds.

Output Delivery

Every command accepts --deliver <sink>. The output goes to the named sink in addition to (or instead of) stdout, so agents can route command results without hand-piping. Three sinks are supported:

Sink Effect
stdout Default; write to stdout only
file:<path> Atomically write output to <path> (tmp + rename)
webhook:<url> POST the output body to the URL (application/json or application/x-ndjson when --compact)

Unknown schemes are refused with a structured error naming the supported set. Webhook failures return non-zero and log the URL + HTTP status on stderr.

Named Profiles

A profile is a saved set of flag values, reused across invocations. Use it when a scheduled agent calls the same command every run with the same configuration - HeyGen's "Beacon" pattern.

posthog-pp-cli profile save briefing --json
posthog-pp-cli --profile briefing flags stale
posthog-pp-cli profile list --json
posthog-pp-cli profile show briefing
posthog-pp-cli profile delete briefing --yes

Explicit flags always win over profile values; profile values win over defaults. agent-context lists all available profiles under available_profiles so introspecting agents discover them at runtime.

Exit Codes

Code Meaning
0 Success
2 Usage error (wrong arguments)
3 Resource not found
4 Authentication required
5 API error (upstream issue)
7 Rate limited (wait and retry)
10 Config error

Argument Parsing

Parse $ARGUMENTS:

  1. Empty, help, or --help → show posthog-pp-cli --help output
  2. Starts with install → ends with mcp → MCP installation; otherwise → see Prerequisites above
  3. Anything else → Direct Use (execute as CLI command with --agent)

MCP Server Installation

Install the MCP binary from this CLI's published public-library entry or pre-built release, then register it:

claude mcp add posthog-pp-mcp -- posthog-pp-mcp

Verify: claude mcp list

Direct Use

  1. Check if installed: which posthog-pp-cli If not found, offer to install (see Prerequisites at the top of this skill).
  2. Match the user query to the best command from the Unique Capabilities and Command Reference above.
  3. Execute with the --agent flag: 
    posthog-pp-cli <command> [subcommand] [args] --agent
  4. If ambiguous, drill into subcommand help: posthog-pp-cli <command> --help.