// SessionEnd hook for routines — turns raw error messages into actionable findings (e.g. "Error: 401" → "Gmail token expired — reconnect at /dashboard/connectors"). Routine-only.
Day-of-week rotating proactive sweep — Mon=connections / Tue=routines / Wed=skills / Thu=memories / Fri=findings / Sat=cross-cutting / Sun=weekly summary. Forces outward scan instead of inward thrash. Routine-only; load this skill into a daily routine that fires AssistantAgent. Adapted from goanna's caretaker.
Use when an agent has had N consecutive quiet runs with nothing to do. Pivots to bounded inward consolidation (promote findings, refresh memory, audit asks) anchored in own data — NOT generic news summarisation. Adapted from goanna's reverie cycle. Routine-only; loaded by the routine that detects the quiet condition.
Use to track open questions you owe the user (asks) and time-bounded commitments you owe yourself (tasks). Write OPEN entries when you commit; promote to CLOSED when answered/done. Survives session compaction. Adapted from goanna's asks.md/tasks.md pattern.
Reviewer criteria for AI-generated code via with_review. Checks for security issues (injection, auth bypass, secret leakage), correctness, and project conventions. Use as criteria.skill when generating code that touches data, auth, or external systems.
Reviewer criteria for outbound email drafts via with_review. Checks tone-recipient match, professionalism, no AI tells, no placeholder leakage. Use as criteria.skill argument when reviewing email drafts before send.
Reviewer criteria for AI-generated summaries via with_review. Checks faithfulness to source, no hallucinated facts, appropriate length, key points preserved. Use as criteria.skill when summarising articles, documents, transcripts, or batch-task outputs.
Raw error messages are useless to non-developer users. Error: fetch failed (status 401) doesn't tell anyone what to do.
This skill rewrites errors into findings with three properties:
For each error, produce an inbox_add payload like:
{
"kind": "routine_error",
"summary": "<plain English what + actionable next step>",
"importance": "medium",
"reasoning": "<technical detail for debugging>",
"suggestedAction": { "label": "<verb-led link text>", "link": "<dashboard path>" }
}
| Raw error | Plain summary | Suggested action |
|---|---|---|
401 Unauthorized from Gmail | "Gmail connection expired" | Reconnect at /dashboard/connectors |
403 Forbidden from Slack | "Slack token doesn't have the scope this routine needs" | Re-authorise in Connectors |
429 Too Many Requests | "Hit rate limit on <service>" | (no action — auto-retries) |
5xx from any service | "<service> is down or having issues" | (no action — wait it out) |
corrupt or unsupported data | "Couldn't process the file" | Check the source file format |
BudgetExceededError | "Daily budget cap reached for this routine" | Raise the cap on the routine detail page |
| Network timeout | "Connection timed out reaching <service>" | (no action — auto-retries) |
If the error is a programming bug (TypeError, ReferenceError, schema mismatch), produce a short routine_error finding tagged for the developer rather than the user, with the raw stack in reasoning. The user can't fix code; the dev can.
Tags to add when relevant: ["dev-bug"] or ["transient"] or ["auth-needed"].