Name: squad-learn
Author: juanjosworkbox

Prompt file imported from juanjosworkbox/hve_laxian (.github/prompts/squad-learn.prompt.md). Fill in {{target}}, {{learning}} before use. Copyright stays with the author.

Squad Learn

Inputs

{{target}}: (Optional) Where to promote the learning. upstream reaches every consumer of the public hve-squad package on their next sync; tenant stays inside your organization's private learnings repository. When omitted, the agent asks after a candidate is drafted.
{{learning}}: (Optional) A specific learning or topic to promote. When omitted, the agent discovers candidates from consumer-local memory.

Requirements

Hand this turn to the Squad Learn agent and let its required steps discover candidates, draft and sanitize the entry, resolve the target repository, and prepare the pull request.
Pass {{target}} and {{learning}} through as-is. The agent owns candidate discovery, the sanitization checklist, and the SL- versus TL- id convention.
Let the agent own its guardrails: live agent memory is read-only, nothing is forked, pushed, or opened without explicit user approval at the impactful-action gate, and unsanitized content stops the run.

# Squad Learn ## Inputs * {{target}}: (Optional) Where to promote the learning. `upstream` reaches every consumer of the public hve-squad package on their next sync; `tenant` stays inside your organization's private learnings repository. When omitted, the agent asks after a candidate is drafted. * {{learning}}: (Optional) A specific learning or topic to promote. When omitted, the agent discovers candidates from consumer-local memory. ## Requirements 1. Hand this turn to the Squad Learn agent and let its required steps discover candidates, draft and sanitize the entry, resolve the target repository, and prepare the pull request. 2. Pass `{{target}}` and `{{learning}}` through as-is. The agent owns candidate discovery, the sanitization checklist, and the `SL-` versus `TL-` id convention. 3. Let the agent own its guardrails: live agent memory is read-only, nothing is forked, pushed, or opened without explicit user approval at the impactful-action gate, and unsanitized content stops the run.

Install squad-learn and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "prompt": "# Squad Learn\n\n## Inputs\n\n* {{target}}: (Optional) Where to promote the learning. `upstream` reaches every consumer of the public hve-squad package on their next sync; `tenant` stays inside your organization's private learnings repository. When omitted, the agent asks after a candidate is drafted.\n* {{learning}}: (Optional) A specific learning or topic to promote. When omitted, the agent discovers candidates from consumer-local memory.\n\n## Requirements\n\n1. Hand this turn to the Squad Learn agent and let its required steps discover candidates, draft and sanitize the entry, resolve the target repository, and prepare the pull request.\n2. Pass `{{target}}` and `{{learning}}` through as-is. The agent owns candidate discovery, the sanitization checklist, and the `SL-` versus `TL-` id convention.\n3. Let the agent own its guardrails: live agent memory is read-only, nothing is forked, pushed, or opened without explicit user approval at the impactful-action gate, and unsanitized content stops the run.", "variables": [ "target", "learning" ], "notes": "Drafts a sanitized, broadly applicable learning from consumer-local squad memory and opens a pull request to promote it, targeting either the public hve-squad package (cross-consumer) or your organization's tenant-internal learnings repository Arguments: [target=upstream|tenant] [learning=...].", "metadata": { "source": { "provider": "github-prompts", "repository": "https://github.com/juanjosworkbox/hve_laxian", "path": ".github/prompts/squad-learn.prompt.md", "ref": "e5b4ae2decbb95306768eaa407b0e496c1f3c738", "url": "https://github.com/juanjosworkbox/hve_laxian/blob/e5b4ae2decbb95306768eaa407b0e496c1f3c738/.github/prompts/squad-learn.prompt.md", "key": "juanjosworkbox/hve_laxian/.github/prompts/squad-learn.prompt.md" } } }

squad-learn

About

Squad Learn

Inputs

Requirements

Use it

Install squad-learn and it becomes one more candidate the router can pick - when it fits.

Manifest

Reviews