Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add anatolykoptev/dozor --skill escalationgit clone --depth 1 https://github.com/anatolykoptev/dozorWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/anatolykoptev/dozor/escalation)<a href="https://agentmods.dev/skills/anatolykoptev/dozor/escalation"><img src="https://agentmods.dev/badge/skills/anatolykoptev/dozor/escalation/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/anatolykoptev/dozor/escalation"><img src="https://agentmods.dev/badge/skills/anatolykoptev/dozor/escalation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00044 | $0.00518 |
| Opus 5 | $0.00022 | $0.00259 |
| Sonnet 5 | $0.00009 | $0.00104 |
| Haiku 4.5 | $0.00004 | $0.00052 |
Grade A, and why
escalation scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Escalation Skill
When you encounter a situation you cannot fully resolve, escalate to the right agent.
Agent Network
| Agent | When to escalate |
|---|---|
| orchestrator | Cross-domain issues, need human decision, status reports, coordination between multiple agents |
| devops | Code bugs, build failures, config changes, complex debugging beyond log analysis |
Process
- DIAGNOSE FIRST — always try to understand the problem yourself (server_triage, server_inspect).
- ATTEMPT FIX — if the fix is safe (restart, cleanup), try it and verify.
- CHOOSE AGENT — pick the right agent based on the table above.
- PROVIDE CONTEXT — include in your a2a_call message:
- What you found (triage results, error logs, metrics)
- What you already tried (restarts, cleanups, their outcomes)
- What you need from them (investigation, decision, fix)
- FOLLOW UP — include the agent's response in your answer to the user.
Examples
Service crash loop → devops:
api-service crashes in a restart loop. Logs show OOM at 512MB.
Already restarted 2x, same result.
Need: memory limit increase or code-level fix.
Disk critical → orchestrator:
Disk at 95%. Ran cleanup, freed 2GB, still at 91%.
Options: prune docker images (~8GB) or expand disk.
Need: decision on which approach.
Cascading failure → orchestrator:
Database, api-service, worker-service all unhealthy.
Root cause: database connection limit reached.
Need: coordinated restart plan, may affect other agents.
Build error → devops:
Deploy failed at build stage.
Error: go build — missing module github.com/example/lib.
Need: code fix or go.mod update.
When NOT to Escalate
- Simple restarts that fix the issue
- Routine health checks with all services healthy
- Cleanup/maintenance tasks within your capabilities
- Single service issues you can diagnose and resolve yourself
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 64 lines · 44 tokens per session scan A 3802057d9aa6
escalation is a skill published in the GitHub repository anatolykoptev/dozor (5 stars, last pushed today), licensed MIT. It adds 44 tokens to every session and 518 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
代码调试
A debugging workflow that reproduces a problem with a failing test, collects evidence, finds the underlying cause, applies a fix, and runs regression checks. Regression checks verify that the fix does not break behavior that already worked.
Godot 编码
A Godot 4 development workflow for editing game scenes and scripts, using the Godot editor to run changes and checking the result. Godot is a tool for building games; scenes describe game objects and their relationships.
System Healthcheck
Run a lightweight, non-destructive system healthcheck — disk usage, memory, CPU load, process count, hostname — and store a structured snapshot in session memory. Use when the user says "run a healthcheck", "check system health", "how's the server doing", "is everything ok", or asks for a status/diagnostics snapshot…
sentry
Sentry error tracking and debugging specialist.
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.
langsmith-observability
LLM observability platform for tracing, evaluation, and monitoring. Use when debugging LLM applications, evaluating model outputs against datasets, monitoring production systems, or building systematic testing pipelines for AI applications.