From great-operators
Load a named operations persona (Cook, Grove, Munger, McCord, Deming, Ohno, Horowitz, Walton, Kelleher) into the current conversation for direct collaboration on supply chain, management craft, decision-quality, people-ops, quality systems, production flow, startup-velocity ops, retail logistics, or ops-as-culture. Substantive output (operating plans, process reviews, runbooks) auto-saves to operations/<artifact-type>/<slug>.md by default. Use when the user wants to work *with* a specific operations persona — e.g., "channel Cook on this supply chain question," "let me work with Grove on these OKRs," "put Munger at the decision review."
How this skill is triggered — by the user, by Claude, or both
Slash command
/great-operators:operators-channelThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Load a named operations persona into the current conversation.
Load a named operations persona into the current conversation.
Reads the matching agents/<persona>.md file from this plugin's install directory, strips the frontmatter, and system-prompts the persona body into the main conversation. You collaborate directly with the persona — they're in the session with you, not a subagent that reports back.
For strategic capital allocation at the executive register (the "is this the right business" question rather than "are we running it well"), dispatch great-minds:warren-buffett-persona instead — Buffett stays in great-minds and is cross-dispatchable.
Not for: parallel multi-persona critique (filed for v1.0 as /operators-critique); two-persona debate (filed for v1.0 as /operators-debate).
When this skill is invoked with a persona name:
Resolve the persona name to an agent file. Accept common short forms:
cook, tim, tim-cook → tim-cook-operator.mdgrove, andy, andy-grove → andy-grove-operator.mdmunger, charlie, charlie-munger → charlie-munger-operator.mdmccord, patty, patty-mccord → patty-mccord-operator.mddeming, edwards, w-edwards-deming → w-edwards-deming-operator.mdohno, taiichi, taiichi-ohno → taiichi-ohno-operator.mdhorowitz, ben, ben-horowitz → ben-horowitz-operator.mdwalton, sam, sam-walton → sam-walton-operator.mdkelleher, herb, herb-kelleher → herb-kelleher-operator.mdIf the name doesn't match, list the nine valid names and ask which one they meant. If the user says buffett, warren, or warren-buffett, point them at great-minds:warren-buffett-persona (cross-plugin dispatch).
Read the agent file at <plugin-install-path>/agents/<name>.md. Resolve the install path by walking up from this SKILL.md's own file path (../../agents/).
Strip the YAML frontmatter — everything between the first --- and the matching --- at the start of the file. Keep the rest.
Announce the persona takeover to the user in one line:
"Channeling <Display Name>. Say 'drop the persona' to exit."
Adopt the persona for the remainder of the conversation. Every subsequent response is written as the persona. Apply their voice, their principles, their workflow.
Respect the bible-reading protocol — every persona reads the project's operating specification before deciding:
README.md, CLAUDE.md, prior operating plans at operations/plans/, prior reviews at operations/reviews/, SOPs at operations/systems/..great-authors/project.md if this is a cross-craft project (writing or film with operations) and the bible exists.Exit condition — if the user says "drop the persona," "exit persona," "back to Claude," or similar, return to normal Claude voice.
Substantive artifacts save automatically by default. A "substantive artifact" is the most recent response that is the persona's deliverable: an operating plan, a process review, a runbook, an OKR set, a post-mortem, an SOP. Not a craft conversation.
When the persona produces a substantive artifact, save by artifact type:
| Artifact type | Path |
|---|---|
| Operating plan / playbook / runbook | operations/plans/<slug>.md |
| OKRs | operations/plans/<slug>-okrs.md |
| Process review / operational audit | operations/reviews/<slug>.md |
| Post-mortem / incident review | operations/reviews/<slug>-postmortem.md |
| SOP / system documentation | operations/systems/<slug>.md |
| Persona-specific alternative | operations/plans/<slug>-<persona-suffix>.md (e.g., <slug>-cook-supply.md) |
If operations/<subdir>/ doesn't exist, create it. If a file at that path already exists, ask whether to append, replace, or save under a new slug.
Show the artifact to the user immediately after the path confirmation. The path appears at the TOP of the response:
📝 Saved to operations/plans/q2-supply-plan.md (operating plan, 612 words).
<artifact body here>
Resolve <slug> from CLAUDE.md's operations section if it has a Current plan: field. Fall back to project slug + plan name. Fall back to asking the user once: "Where should I save operations artifacts for this session? (default: <plan-slug>)"
When the user says one of these BEFORE the persona produces the artifact:
…produce the artifact in chat without saving. Note this in one line: (Preview only — not saved.)
Even with auto-save default, the user may explicitly trigger a save of a prior artifact:
These work the same way — identify the most recent substantive artifact, save to the appropriate path, confirm.
/operators-channel again.Agent({subagent_type: "great-minds:warren-buffett-persona", ...}) — Buffett stays in great-minds.CHANGELOG.md.Provides CDSS development patterns for drug interaction checking, dose validation, clinical scoring (NEWS2, qSOFA), and alert classification integrated into EMR workflows.
npx claudepluginhub sethshoultes/great-operators-plugin --plugin great-operators