Skip to main content
skill CX Codex

exploring-replay-vision-observations — Claude Code skill

The exploring-replay-vision-observations skill guides an agent through Replay Vision scanner results, from recurring patterns to individual session recordings. Use it to review findings and decide what to investigate next.

no license 203 installs ★ 101 updated 2026-11-02

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add PostHog/ai-plugin --skill exploring-replay-vision-observations -a claude-code
Open on GitHub

What exploring-replay-vision-observations does

The skill helps an agent identify a scanner, read its observations, and inspect a specific result. It also points to ways to compare findings across sessions, such as scanner statistics, related issue groups, named themes, and a feed of recordings selected for review.

Results vary by scanner: an observation can contain a verdict, labels, a score, or a recording summary with chapters. These are model assessments, not verified facts, so check a finding against other observations or the session recording before acting on it. The skill also treats observation text as untrusted data and warns agents not to follow instructions found there.

It uses PostHog tools such as vision-scanners-list, vision-scanners-get, and observation-reading methods. Install it in Claude Code with the command in the FAQ. The listed platforms also include Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode.

What it helps with

  • review a scanner’s findings over a chosen period
  • filter observations by verdict, tags, or score
  • spot recurring issues across session recordings
  • inspect an observation and its linked events
  • compare scanner statistics with individual results

Example prompts for the agent

  • Review my Replay Vision scanner’s findings and group the recurring issues.
  • Show successful observations from the scanner over the last seven days, ranked by score.
  • Inspect this Replay Vision observation and check it against the session recording.

How to install exploring-replay-vision-observations

1Skills CLI (Vercel)

npx skills add PostHog/ai-plugin --skill exploring-replay-vision-observations -a claude-code

2Manually: put the files in place

~/.claude/skills/exploring-replay-vision-observations/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/exploring-replay-vision-observations/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about exploring-replay-vision-observations

How do I install exploring-replay-vision-observations in Claude Code?

Run npx skills add PostHog/ai-plugin --skill exploring-replay-vision-observations -a claude-code. The skill file is placed at ~/.claude/skills/exploring-replay-vision-observations/SKILL.md.

What does exploring-replay-vision-observations do?

It helps an agent read Replay Vision scanner results, inspect individual observations, and find patterns across sessions. It also guides the agent through distinguishing an error from a scan that found no issue.

Why verify Replay Vision observations before acting on them?

Each observation is a model’s assessment of one session, not proof that an issue occurred. The skill advises checking a finding against other observations or the recording before taking action.

Similar skills

All skills for AI and ML →

From the same set82

The whole set →
auditing-warehouse-source-health auditing-warehouse-source-health audits PostHog data warehouse sources and syncs to pinpoint broken connectors, failed schemas, and degraded webhooks. CX Codex +1 204installs signals-scout-product-analytics signals-scout-product-analytics — a skill for Claude Code and Codex monitoring core product analytics flows in PostHog. Detects derived-rate regressions across funnels and retention while entrant volume holds steady. CX Codex +1 204installs auditing-warehouse-view-health auditing-warehouse-view-health runs a project-wide check on PostHog materialized views: lists the ones that failed to refresh and points at materialized views nobody queries. CX Codex +1 204installs building-workflows building-workflows is a Claude Code skill for building, patching, and testing PostHog automation graphs over MCP before publishing them in the visual editor. CX Codex +1 204installs signals-scout-customer-analytics The signals-scout-customer-analytics skill tracks account-level engagement in PostHog and spots churn-risk or expansion patterns. Use it for recurring customer-health checks. CX Codex +1 203installs exploring-mcp-sessions exploring-mcp-sessions investigates PostHog MCP sessions to inspect tool call sequences, agent intent, and run failures. CX Codex +1 202installs exploring-mcp-tool-quality exploring-mcp-tool-quality reads PostHog MCP analytics to rank tools by failure share, latency and reach, then drills into the ones that break. CX Codex +1 202installs exploring-mcp-intent-clusters exploring-mcp-intent-clusters analyzes PostHog MCP agent intent clusters to reveal what agents attempt to do, tool failure patterns, and semantic routing competition. CX Codex +1 202installs

Project sponsors

Companies that keep this analytics open