CtrlK
BlogDocsLog inGet started
Tessl Logo
PluginPublic

ainativedev/aidevcon-2026-nyc

AI Native DevCon 2026 New York — practical agent skills grounded in conference talks

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Overview
Quality
Evals
Security
Files

application-template.mdskills/talk-stack-incident-dossiers/references/

Incident dossier and next decision

This is a recommended format applying Stack's workflow. It is not authorization to remediate or suppress an alert.

Incident identity

Title, category and affected service: State what is actually known.

Trigger and observation window: Record the alert source, first/last seen times and evidence of provenance. Mark signature verification unknown unless inspected.

Correlation: List related incidents and repeated notifications. Explain whether the symptoms are unchanged or new; separate model classification from the rule deciding incident state.

Findings and evidence

FindingSupporting evidenceTime/scopeConfidence or limitation
Observed factQuery result, log or supplied recordActual windowMissing or conflicting evidence

Include deployments, configuration, dependencies and runbook checks relevant to this alert. Explicitly list checks unavailable or not completed.

Likely causes

Rank plausible explanations only as far as evidence permits. For each, give supporting evidence, a competing explanation and the next discriminating check. Do not label an agent hypothesis a confirmed root cause.

Operator decision

Describe recommended next steps, their operational consequences and what additional judgment is needed. State what has and has not been changed.

Use the team's existing incident and permission process. A read-only dossier does not imply that a rollback, certificate rotation or page suppression is authorized.

Follow-up and learning

Record the operator's question, investigation performed and answer in the incident history. Preserve source timing and scope.

After resolution, capture one workflow improvement: a missing check, a clearer field, a correlation rule or an access fix. Define a representative replay and expected result, marking it planned until performed.

End with a concise decision-ready summary, unresolved uncertainty and the next responsible action. Keep the layout stable across incidents so operators can find evidence quickly.

skills

talk-stack-incident-dossiers

tile.json