agentleFS
HybridAIOneHybridAIOne/hybridclaw/skills/posthog/SKILL.md

posthog

Skill159 starsChanged 31 days agoMITView on GitHub

Capture PostHog events, update person properties, read persons, inspect feature flags, test flag evaluation, and run bounded analytics queries through gateway-managed secrets.

  • Reads credentials

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. 0 are common, 9 are distinctive, and 7 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 9

    Distinctive

    This author's own choices. Worth a look.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill for PostHog product analytics work: event capture, person property reads and updates, feature flag inspection, feature flag …

    line 84
  2. Very specific:

    update person properties with a guarded $identify capture event

    line 91
  3. Distinctive:

    test how a feature flag evaluates for a distinct id without changing the flag

    line 94
  4. Distinctive:

    run private query API requests for HogQL, trends, funnels, retention, and other PostHog query payloads

    line 95
  5. Very specific:

    Never paste either token into chat or helper arguments.

    line 116
  6. Very specific:

    Use the right regional hosts.

    line 147
  7. Very specific:

    Use plan for natural-language requests when you need to classify read vs write risk before executing.

    line 155
  8. Distinctive:

    Run the bundled helper for live PostHog calls so request construction, gateway submission, credentials, and error interpretation stay in …

    line 157
  9. Very specific:

    Use http-request only when you need to inspect the generated request or when a runtime exposes the built-in http_request tool but cannot …

    line 162
  10. Distinctive:

    Keep capture payloads small and business-relevant.

    line 172
  11. Distinctive:

    If a live PostHog call returns 401 or 403, stop after that first failure and ask the operator to verify the matching stored credential and …

    line 175
  12. Very specific:

    If a private response is paginated and includes next, call the next URL only when the user needs another page.

    line 177
  13. Distinctive:

    Update person properties after explicit approval:

    line 220
  14. Distinctive:

    Read feature flags and test evaluation:

    line 242
  15. Distinctive:

    Run an analytics query:

    line 255
  16. Very specific:

    Use --query-json for PostHog query payloads beyond HogQL:

    line 262

“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

PostHog

Use this skill for PostHog product analytics work: event capture, person property reads and updates, feature flag inspection, feature flag test evaluation, and bounded insight/HogQL queries.

Scope

  • capture a single product analytics event through the public capture endpoint
  • update person properties with a guarded $identify capture event
  • list and retrieve persons through the private persons API
  • list and retrieve feature flag metadata
  • test how a feature flag evaluates for a distinct id without changing the flag
  • run private query API requests for HogQL, trends, funnels, retention, and other PostHog query payloads
  • classify common PostHog auth, permission, validation, and rate-limit errors
  • measure skill run cost through normal HybridClaw UsageTotals

Out Of Scope

  • creating, editing, rolling out, or deleting feature flags
  • deleting persons or bulk deleting data
  • bulk historical imports or migration-sized batch capture
  • exporting large event/person tables on a schedule
  • bypassing PostHog region, project, environment, or credential configuration

Credential Rules

PostHog has two credential rails:

  • POSTHOG_PROJECT_TOKEN is the public project token used in capture payloads.
  • POSTHOG_PERSONAL_API_KEY is a private bearer credential used for persons, feature flags, and query endpoints.

Never paste either token into chat or helper arguments. The helper emits <secret:POSTHOG_PROJECT_TOKEN> inside capture JSON and bearerSecretName: "POSTHOG_PERSONAL_API_KEY" for private APIs, so the gateway injects credentials server-side.

Recommended setup order:

  1. Browser admin: open the active HybridClaw admin URL ending in /admin/secrets and set POSTHOG_PROJECT_TOKEN and POSTHOG_PERSONAL_API_KEY.
  2. Browser /chat or TUI fallback:
/secret set POSTHOG_PROJECT_TOKEN <project-token>
/secret set POSTHOG_PERSONAL_API_KEY <personal-api-key>
/env set POSTHOG_HOST "https://us.posthog.com"
/env set POSTHOG_INGEST_HOST "https://us.i.posthog.com"
/env set POSTHOG_PROJECT_ID "12345"
/env set POSTHOG_ENVIRONMENT_ID "12345"
  1. Local console fallback:
hybridclaw secret set POSTHOG_PROJECT_TOKEN "<project-token>"
hybridclaw secret set POSTHOG_PERSONAL_API_KEY "<personal-api-key>"
hybridclaw env set POSTHOG_HOST "https://us.posthog.com"
hybridclaw env set POSTHOG_INGEST_HOST "https://us.i.posthog.com"
hybridclaw env set POSTHOG_PROJECT_ID "12345"
hybridclaw env set POSTHOG_ENVIRONMENT_ID "12345"

Use the right regional hosts. For PostHog US Cloud, private API calls use https://us.posthog.com and capture calls use https://us.i.posthog.com. For EU Cloud, use https://eu.posthog.com and https://eu.i.posthog.com. For self-hosted deployments, use the self-hosted base URL for both when that is how the instance is exposed.

Default Workflow

  1. Use plan for natural-language requests when you need to classify read vs write risk before executing.
  2. Run the bundled helper for live PostHog calls so request construction, gateway submission, credentials, and error interpretation stay in one place:
    node skills/posthog/posthog.cjs --format json run ...
  3. Use http-request only when you need to inspect the generated request or when a runtime exposes the built-in http_request tool but cannot run the helper against the gateway directly:
    node skills/posthog/posthog.cjs --format json http-request ...
  4. Pass only the emitted httpRequest object to the built-in http_request tool in that fallback path. Do not handcraft PostHog API calls from memory.
  5. For amber operations, run approval-plan first, get explicit operator confirmation, then rerun the exact helper command with --operator-grant.
  6. Keep capture payloads small and business-relevant. Do not send passwords, access tokens, full message bodies, contracts, or raw support transcripts as event/person properties.
  7. If a live PostHog call returns 401 or 403, stop after that first failure and ask the operator to verify the matching stored credential and scopes.
  8. If a private response is paginated and includes next, call the next URL only when the user needs another page.

Command Contract

Inspect the helper surface:

node skills/posthog/posthog.cjs --help

Plan a request without contacting PostHog:

node skills/posthog/posthog.cjs --format json plan "Show active flags for checkout"
node skills/posthog/posthog.cjs --format json plan "Capture a trial_started event for user_123"

Build an approval plan for a capture write:

node skills/posthog/posthog.cjs --format json approval-plan capture-event \
  --event trial_started \
  --distinct-id user_123 \
  --properties-json '{"plan":"pro"}'

Capture a single event after explicit approval:

node skills/posthog/posthog.cjs --format json run capture-event \
  --event trial_started \
  --distinct-id user_123 \
  --properties-json '{"plan":"pro"}' \
  --operator-grant

node skills/posthog/posthog.cjs --format json http-request capture-event \
  --event trial_started \
  --distinct-id user_123 \
  --properties-json '{"plan":"pro"}' \
  --operator-grant

Update person properties after explicit approval:

node skills/posthog/posthog.cjs --format json http-request identify-person \
  --distinct-id user_123 \
  --set-json '{"company":"Acme GmbH","plan":"pro"}' \
  --operator-grant

Read persons:

node skills/posthog/posthog.cjs --format json run list-persons \
  --environment-id 12345 \
  --search acme \
  --limit 50

node skills/posthog/posthog.cjs --format json http-request get-person \
  --environment-id 12345 \
  --person-id 018f6c8f-...

Read feature flags and test evaluation:

node skills/posthog/posthog.cjs --format json http-request list-feature-flags

node skills/posthog/posthog.cjs --format json http-request get-feature-flag \
  --flag-id 42

node skills/posthog/posthog.cjs --format json http-request test-feature-flag \
  --flag-id 42 \
  --distinct-id user_123

Run an analytics query:

node skills/posthog/posthog.cjs --format json http-request query \
  --hogql "select event, count() from events where timestamp > now() - interval 7 day group by event order by count() desc limit 10"

Use --query-json for PostHog query payloads beyond HogQL:

node skills/posthog/posthog.cjs --format json http-request query \
  --query-json '{"kind":"TrendsQuery","series":[{"kind":"EventsNode","event":"$pageview"}]}'

Interpret a saved http_request error:

node skills/posthog/posthog.cjs --format json explain-error --payload-file /tmp/posthog-error.json

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.