Overview
Type/insight in the input box of any AI SRE session. AI SRE reviews your sessions from the last 30 days and produces a single-page operational insight report. The report uses a parchment-style HTML layout and renders directly in the session as a chat card, which you can preview and download.
It is a coach’s read on the month, not a problem list. It leads with what’s working — where AI SRE genuinely pulled its weight this month. Then it covers the frictions worth eliminating now: patterns that keep draining your time. Examples include pasting the same database connection string into several sessions, an agent missing a runbook it should know, or an agent querying the wrong data source again and again. And finally offers 2–3 higher-leverage next steps — shifts in how you work with AI SRE that are worth trying.
/insight is read-only. It only analyzes and presents; it will not automatically modify any knowledge, skill, or MCP configuration. Every suggestion is copyable text — whether to act on it is entirely up to you. See How to Act on Suggestions.How to generate
1
Type /insight in a session
Type
/insight in the input box of any AI SRE session and send it. No parameters are required.2
AI SRE automatically determines the analysis scope
The scope depends on whether the current session is bound to a team (see the table below). AI SRE will state the analysis scope and time window (default: last 30 days) at the beginning of the report.
3
Wait for the report to render
AI SRE analyzes those sessions and then renders the parchment-style report as a chat card, accompanied by a 2–3 sentence spoken summary. You can preview or download the full report from the card.
Analysis scope
/insight always uses the account as the security boundary — it only analyzes sessions you are allowed to see. Within that boundary, the specific scope depends on whether the session is bound to a team:
How the report is produced
AI SRE takes the in-scope sessions from the last 30 days (up to 200 by default, across web, IM, API, and automation) and keeps only sessions with real signal: ≥ 2 user-message turns or ≥ 3 tool calls (either qualifies). That way, autonomous runs started by an alert or a schedule — no human turns, but real investigation — are included too. Every conclusion carries a verbatim quote as evidence. The report shows no session IDs and does not copy full raw session content.Report contents
The report has six parts from top to bottom. Regardless of whether any friction is found, the quantitative overview is always present.1. At a glance
A 2–4 sentence coaching synthesis at the top of the report. AI SRE writes it last and reduces the month to the gist: volume and top theme, the one thing working best, and the one friction to fix first.2. Overview (quantitative)
Includes:3. Narrative
A 2–4 sentence second-person summary of your main work this month. It covers the most common domains, recurring entities, and the overall shape of the month. Recurring entities include an incident analyzed more than once or a cluster or host that kept appearing. This part is aggregated from the topics surfaced across your sessions.4. What’s working
The top ~3 concrete, evidence-backed things AI SRE did well — cross-source validation, correct noise triage, a clean root-cause analysis. This part balances the report and is never skipped when there are real wins; each carries a verbatim quote as its evidence.5. Frictions
Friction cards ranked from highest to lowest importance, up to approximately 8 cards. Each card includes:- A rank and a friction type label (one of five types; see the next section);
- A frequency badge showing how many distinct sessions this friction consumed your time in (deduplicated evidence session count);
- A one-line title and a second-person explanation;
- Evidence: a single verbatim quote from the transcript (the most telling user or agent line);
- A copyable recommendation: text you can paste directly, whose first line (
Add this to knowledge/<scope>/…) names the target file.
6. Next steps
About 2–3 forward-looking, grounded suggestions. These are strategic (a shift in how you work with AI SRE, higher-leverage), distinct from the tactical, file-level fixes in the friction cards. Each names the exact observation it’s grounded in (a stat from the overview or a friction cluster) and maps to a real AI SRE capability. Two common kinds of next step:- Move an automation’s collection and aggregation into a tested script: an automation session might pull raw listings into context and have the agent count, group, or rank them itself. If so, ask the agent in chat to fix it. The agent moves the collection and aggregation into a tested script under
knowledge/<scope>/automations/<slug>/and leaves the task prompt with only the goal, the delivery, and a pointer to that folder. - Add a datasource: sessions can expose a root-cause gap on the datasource side: the same type in ≥ 2 sessions, or explicit evidence in one. The report then suggests that you add the matching datasource in Monitors (Prometheus, Loki, VictoriaLogs, SLS, Tencent Cloud CLS, Elasticsearch, MySQL, PostgreSQL, Oracle, ClickHouse). After you add it, AI SRE can query it automatically, with no skill to install. The report never creates the datasource for you.
Every friction and win must be grounded in at least one real session and carry a verbatim quote as evidence — the report never fabricates sessions, facts, or runbook gaps. If no friction is found, the frictions part shows an empty-state message and the rest of the report still renders.
Friction types
/insight recognizes exactly five friction types, each with a clear place to fix it. Listed in default importance order:
How to act on suggestions
The report only provides copyable fix text; you apply it yourself in the corresponding resource.1
Read the report and pick what's worth doing
Friction cards are already ranked by importance. Start with the top-ranked cards that have high frequency badge numbers — those are the patterns consuming your time most repeatedly.
2
Copy the suggested text
Copy the card’s “Copyable suggestion” text; its first line (
Add this to knowledge/<scope>/…) names the target file.3
Manually apply it in the corresponding resource
Based on the friction type, paste and save in the appropriate resource: repeated context and hallucinated entities go into the knowledge’s
services.md or DUTY.md. A missing runbook means adding a new runbook to the knowledge. A wrong data source gets pinned in observability.md / clusters.md. Stale knowledge means updating the outdated file directly. These are all standard edits under Knowledge./insight runs only when you type it; it does not generate reports on a schedule. Run it again after a few more investigations to see which frictions are gone and which are new.Related pages
Console Chat
Learn about session creation, team binding, and war rooms — these determine the analysis scope of
/insight.Manage Knowledge
Most suggestions from the report land in the knowledge’s DUTY.md /
services.md / runbooks. This page explains how to edit them.Overview
Get a high-level understanding of AI SRE’s capabilities and how it works.
Artifacts
A report generated by
/insight is itself an artifact — once published, view and share it from the artifact gallery.