fix(skill): valid YAML + ≤1024 description for ornn-agent-manual-cli frontmatter

.changeset/fix-manual-cli-frontmatter-yaml.md

@@ -0,0 +1,4 @@
+---
+---
+
+docs(skill): fix `ornn-agent-manual-cli` v1.4 frontmatter — the `description` contained an unquoted `: ` (colon-space) that broke YAML parsing on ingest/refresh, and exceeded the 1024-char cap. Reworded to remove the colon and trimmed to 993 chars; verified against the API's `yaml.parse` + `validateSkillFrontmatter`. Docs-only.

skills/ornn-agent-manual-cli/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ornn-agent-manual-cli
-description: The manual an AI agent loads to operate Ornn, the model-agnostic skill-lifecycle API (an npm-style registry + CLI for agent skills), through the NyxID CLI (`nyxid proxy request ornn-api …`). Load and follow this skill WHENEVER the user asks to work with Ornn skills or skillsets, e.g. search Ornn / find a skill, pull or install a skill or a specific version, run or execute a skill, build and upload a new skill, publish a new version, make a skill public / private / shared, run a security audit or read audit history, deprecate or delete a skill or version, diff two versions, check usage analytics, bind a skill to a NyxID service, link a skill to GitHub or sync from source, or check your quota / pick an LLM model before an SSE call — and for every SKILLSET request: bundle skills into a set, create or publish a skillset (a curated multi-skill package with a required master prompt), resolve a skillset closure in one call, export a skillset as a Claude Code marketplace plugin, transfer skill or skillset ownership, or work out why others can't see a skillset you shared (visibility is derived from member skills). Once loaded the agent runs the whole search → pull → execute → build → upload → share lifecycle with no further setup. Authoritative Ornn↔agent contract; pair with references/api-reference.md (the full per-endpoint catalogue + error legend), shipped together as one Ornn skill.
+description: The manual an AI agent loads to operate Ornn — the model-agnostic skill-lifecycle API (an npm-style registry + CLI for agent skills) — via the NyxID CLI (`nyxid proxy request ornn-api …`). Load and follow it WHENEVER the user asks to work with Ornn skills or skillsets, e.g. search Ornn or find a skill, pull or install a skill or a specific version, run a skill, build and upload one, publish a new version, make a skill public/private/shared, run or read a security audit, deprecate or delete, bind a skill to a NyxID service, link to GitHub or sync from source, check quota or pick an LLM model; and for skillsets, bundle skills into a set, create or publish a skillset (a curated multi-skill package with a master prompt), resolve its closure in one call, export it as a Claude Code marketplace plugin, transfer ownership, or work out why others can't see a skillset you shared (visibility is derived from members). Authoritative Ornn↔agent contract; pair with references/api-reference.md.
 metadata:
   category: plain
   tag: