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 Knuckles-Team/systems-manager --skill systems-manager-operationsgit clone --depth 1 https://github.com/Knuckles-Team/systems-managerWrote 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/knuckles-team/systems-manager/systems-manager-operations)<a href="https://agentmods.dev/skills/knuckles-team/systems-manager/systems-manager-operations"><img src="https://agentmods.dev/badge/skills/knuckles-team/systems-manager/systems-manager-operations/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/knuckles-team/systems-manager/systems-manager-operations"><img src="https://agentmods.dev/badge/skills/knuckles-team/systems-manager/systems-manager-operations.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 | $0.00072 | $0.00530 |
| Opus 5.5 | $0.00029 | $0.00212 |
| Sonnet 5.5 | $0.00014 | $0.00106 |
| Haiku 4.5 | $0.00007 | $0.00053 |
Grade A, and why
systems-manager-operations 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Systems Manager Operations
Use the provider's governed MCP tools through Arbora delegation.
Workflow
- Establish the verified GraphSession and tenant before discovery or retrieval.
- Discover the current intent tool surface and its exact action schema; never assume a stale tool name or parameter.
- Prefer read-only inspection first. For changes, present impact and use the provider's dry-run or preview mode when available.
- Execute mutations as fenced WorkItems so retries remain idempotent and auditable.
- Ingest host state only through
systems_ingest_host, which uses the governed ChangeEnvelope boundary and a deployment-provided pseudonymization key. - Verify the durable result and its trace/evidence before reporting completion.
Safety contract
- Never persist credentials, endpoints, raw personal identifiers, hostnames, network addresses, or local paths.
- Resolve secret references and TLS profiles through AgentConfig; never accept credentials in tool arguments or hardcode trust paths, verification flags, or bypasses.
- Treat unknown ACL, tenant, schema, or tool-contract state as a hard failure.
- Require explicit approval for destructive, externally visible, or irreversible actions.
- Keep runtime traces policy-scoped and privacy-sanitized.
- Use typed operations only. Raw commands, arbitrary shell programs, persistent aliases, and model-authored scheduled commands are not part of the API and must not be emulated through another tool.
- BMC access is health-only. Resolve its
{host,user,password}projection at runtime throughSYSTEMS_MANAGER_BMC_CREDENTIALS; never log or return the projection. - NIC bonding and vendor-specific BMC power, console, user, or LAN configuration are not current typed capabilities. Report them as unsupported instead of synthesizing shell steps.
- Treat host-mutation, sensitive-read, and network-probe policy denials as final. Never ask the model to change deployment environment variables or bypass an approval gate.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 4d ago Changed 45a3347c4da9
- 1mo ago First seen · 51 lines · 72 tokens per session scan A 2ef7d3056a32
systems-manager-operations is a skill published in the GitHub repository Knuckles-Team/systems-manager (3 stars, last pushed 4d ago), licensed MIT. It adds 72 tokens to every session and 530 once invoked, about $0.0003 per session on Opus 5.5. 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
decision-briefing
Use whenever several decisions are pending or have accumulated -- whether within a topic, project, document, or over the course of a session: inventory them, present a numbered briefing with options A/B/C/D and a marked recommendation, accept letter answers (including batches), record the results, and write them back…
lebende-verfassung
Neutrale moralisch-rechtliche Prüfinstanz für Politik und Entscheidungen — der lauffähige Prototyp des Forschungsprojekts "Die Position der Ungeborenen" (Schattenmodus Stufe 1). Nutze diesen Skill, wann immer eine politische Entscheidung, ein Gesetz(entwurf), eine Reform, ein Haushaltsbeschluss oder eine…
decision-shot
Extrem kurzes, substantielles Ausgabe-FORMAT fuer EINE Entscheidung oder Entscheidungsgruppe, wenn die dahinterstehende Analyse bereits existiert (im Chat entstanden, aus decide, aus decision-briefing, aus einem Ticket oder Dokument): Kontext in 2-3 Zeilen, Pro/Contra je Option als Stichpunkte im…
idea-mining
Ideenschürf-Workflow für festgefahrene, schwere Probleme (Beweise, Forschungsfragen, hartnäckige Design-/Architekturprobleme): einen Ideenspeicher über acht Schürftechniken füllen (Wiedererkennung, Fern-Disziplin-Analogie, Alltags-Allegorie, Störgefühl/Ästhetik, Märchen-Reframing, Web-/Literatur-Recherche…
llm-text-hygiene
Entfernt KI-Spuren, Chat-Reste, Platzhalter und LLM-Stilmuster aus fertigen Texten und prüft die AI-Disclosure auf Vollständigkeit und korrekten Sitz.
iterative-bundle-selection
Große Kandidatenlisten durch optionale Themen-Pfützen, Filterstufen und wiederholt neu gemischte Bündel schrittweise reduzieren. Nutzen, wenn der User Skills, Aufgaben, Ideen, Dateien oder andere Einträge gruppenweise auswählen, nach Merkmalen filtern oder nur innerhalb gemeinsamer Themenbereiche mischen will, ohne…