Index of every page on this site: llms.txt
All skills
launchdarkly avatar

/alert-investigation

@f0ee110 official
Run it yourself

Investigates a firing or recently fired observability alert: finds the firing and replays the value it evaluated, decides whether the alert is chronic noise or a new problem, reuses any automatic investigation, then finds the cause (what changed, which host, version or route is over-represented) and returns a structured diagnosis with cause, scope and next steps.

Use this Skill: https://skilld.dev/gh/launchdarkly/ai-tooling/alert-investigation

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

sessions.md

≈545 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Alert investigation — sessions

Load this file when investigating a session-based alert (session-error rate, user-facing failure, session-quality anomaly).

Before you look for a cause

  • Replay the firing with the replay arguments from get-alert-firing-history, then with compareTo: "1w". Session counts follow traffic; a rise that matches last week's shape is not an incident.

Investigation shape

  1. Pull matching sessions via search-sessions with the alert's query (from get-alerts with alertId) and the firing window. Start small — 5–10 sessions sampled from the window.
  2. For each session, walk the timeline via search-timeline-events. Find the last successful interaction and the first failing one. The transition is the key moment.
  3. Cross-reference errors. If sessions have has_errors=true, pull the error group for each. If sessions share an error group, that's the common root.
  4. Correlate backend and frontend. Session events often carry trace IDs — follow them to backend traces and logs for the full picture.
  5. Check for a common trigger. Is it a specific browser? Geography? Time of day? Shared identifier attribute?

What goes in the diagnosis

  • What triggered — alert name, threshold, the query pattern matching the affected sessions.
  • Likely cause — the common failure point across sessions with one quoted session ID as example. Cite specific timeline events.
  • Scope — number of affected users, geographic or demographic clustering, time-window pattern.
  • Next steps — concrete: "roll back the <feature> deploy", "disable flag <key>", "check for a degraded third-party script on the target page".

Wrap session IDs

Whenever you reference session IDs in the diagnosis, wrap them in <session-id> tags:

<session-id>c6wX2TtznM6F0QByEiRBoQBFw1Bv</session-id>

The UI renders these as clickable links.

Common mistakes

  • Skipping the timeline walk and inferring behavior from session attributes alone.
  • Paraphrasing errors instead of pulling the error group record.
  • Forgetting to wrap session IDs in <session-id> tags. The UI can't link them otherwise.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warningtoday3 checks · Risk SAFE
  • Gen Agent Trust Hubtoday

    The skill is designed to investigate observability alerts by querying a LaunchDarkly MCP server. It retrieves logs, traces, metrics, and session data to diagnose system failures. The primary security consideration is the ingestion of untrusted data from logs and error messages, which presents a surface for indirect prompt injection.

  • Sockettoday

    No alerts

  • Snyktoday

    Risk: MEDIUM · 2 issues

Signed by skilld at f0ee110. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 days ago
compatibility
Requires the remotely hosted LaunchDarkly MCP server
metadata
{
  "author": "launchdarkly",
  "version": "0.3.0"
}

README badge

README badge for launchdarkly/ai-tooling/alert-investigation