agentleFS
MicrosoftKnown publishermicrosoft/sre-agent/labs/zava-learning/sre-config/agent-config/skills/pagerduty-incident-update/SKILL.md

pagerduty-incident-update

Skill190 starsChanged 6 months agoMITView on GitHub

Use to keep the triggering PagerDuty incident in sync with the investigation and to close it out — acknowledge on engagement, mark the incident Mitigated the moment the live fix lands, post a concise symptom-only status/summary note (diagnosis, mitigation, links to the PR / Change Request), and resolve the incident only after the RCA and summary report are complete and recovery is verified. This is the PagerDuty communication and closure layer for a Zava Learning incident.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Keep the PagerDuty incident that triggered this investigation current, and close it out cleanly.

    line 17
  2. Very specific:

    Use AddNoteToPagerDutyIncident to add a concise, factual note.

    line 41
  3. Distinctive:

    Add interim notes at meaningful transitions (mitigation applied, recovery confirmed) rather than a single dump at the end.

    line 54
  4. Very specific:

    Call ResolvePagerDutyIncident to close the incident only once the incident is already MITIGATED, recovery is proven, AND the RCA + summary …

    line 58
  5. Very specific:

    No secrets, credentials, or PII in notes.

    line 68
  6. Distinctive:

    Don't resolve before recovery is verified.

    line 71

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “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

Zava Learning — PagerDuty Incident Update & Closure

Keep the PagerDuty incident that triggered this investigation current, and close it out cleanly. Resource Group: @@RG@@. Services: learner-portal, course-api, assessment-api. Operate on the PagerDuty incident id from the triggering event (confirm details with GetPagerDutyIncidentById if needed). Capture the incident's html_url from the GetPagerDutyIncidentById response — that is the operator-facing PagerDuty link the zava-reporting Artifacts & links section surfaces as [Incident #<n>](<html_url>). The bare incident key (e.g. Q1LXN8LVWLS9L6) is NOT clickable — always pass the html_url.

Acknowledge on engagement

As soon as you start working the incident, call AcknowledgePagerDutyIncident so responders see it is owned and stop escalating. This must be a REAL tool call that returns success — do not write "acknowledged" unless AcknowledgePagerDutyIncident actually executed in this thread. If that tool is not available, this skill is not loaded: load pagerduty-incident-update and retry before doing anything else.

Mark MITIGATED immediately after the live fix

The moment the live mitigation is applied and the symptom clears (endpoint returns 200), post a plain-text note with Status: Mitigated via AddNoteToPagerDutyIncident, summarising the live fix in one or two lines. This marks the incident as mitigated for on-call while the incident stays acknowledged — do NOT resolve here. The lifecycle is always: mitigate → mark Mitigated → RCA + summary → resolve. (If zava-reporting or a triage skill swapped the active toolset and the PagerDuty tools are gone, reload pagerduty-incident-update first.)

Post the status / summary note

Use AddNoteToPagerDutyIncident to add a concise, factual note. PagerDuty notes are PLAIN TEXT only — they do not render HTML or markdown, and they have a practical length cap. Never paste the zava-reporting HTML email, the deck, or the Teams Adaptive Card into a note; those rich deliverables go through their own channels. Instead post a short text summary and link out to the rich artifacts. Keep it symptom-only — never put the root cause in a heading or opening line (the alert is symptom-only by design). Pull the agreed summary from the completed rca-analysis (and zava-reporting) via SearchMemory; do not re-derive it. Post (plain text):

  • Symptom & impact: one line on what users experienced and the blast radius.
  • What we found / did: the confirmed diagnosis and the live mitigation applied, in plain text.
  • Links: the GitHub pull request (durable fix), the ServiceNow Change Request, and the full report/deck location if available.
  • Status: Mitigated / Monitoring / Resolved.

Add interim notes at meaningful transitions (mitigation applied, recovery confirmed) rather than a single dump at the end.

Resolve — only after mitigation, RCA, and the summary report are all complete

Call ResolvePagerDutyIncident to close the incident only once the incident is already MITIGATED, recovery is proven, AND the RCA + summary report exist: the public endpoint returns 200 on / and /api/quiz/*, the underlying Azure Monitor alert has auto-mitigated, evidence-before-after confirms the metric/state returned to healthy, and the RCA narrative and Zava report have been produced. Never resolve straight after the live fix — that step only marks the incident Mitigated; resolution comes after the RCA and summary. Before resolving, ensure a final note records the resolution and references the PR and Change Request. Do not resolve on a hunch or while the alert is still firing.

Rules

  • Symptom-only language in every note title/opening — root cause goes in the body, never a heading.
  • No secrets, credentials, or PII in notes. Apply the zava-redaction standard (SearchMemory("zava-redaction")) and mask anything sensitive as [REDACTED:<CLASS>] before posting — never paste a Key Vault secret value, connection string, or token into a note.
  • Don't resolve before recovery is verified; if recovery is partial or unconfirmed, leave it acknowledged with a Monitoring note and say why.
  • Blameless: describe systems and conditions, not individuals.

Verification

The PagerDuty incident is acknowledged on engagement, marked Mitigated the moment the live fix lands (a Status: Mitigated note), carries a concise symptom-only summary note with links to the PR and Change Request, and is resolved only after the RCA and the Zava report are complete and recovery is verified — i.e. the lifecycle ran mitigate → mark Mitigated → RCA + summary → resolve.

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.