oh-my-knowledge 0.31.0 → 0.32.0

package/docs/prompts/llm-enhanced-review.prompt.md

@@ -1,111 +0,0 @@
-# LLM Enhanced Review Prompt
-
-promptId: llm-enhanced-review
-promptVersion: 2026-05-18.v1
-
-You are reviewing one skill runtime chain from an evidence pack. The deterministic pipeline already extracted facts. Your job is to add semantic review, not to replace raw evidence.
-
-Rules:
-- Return only one valid JSON object.
-- Do not invent evidence. Every judgment or suggestion must be grounded in the provided skill content or runtime summary.
-- `userGoal` must describe the concrete runtime user goal from `runtimeEvidence.userMessages` / `runtimeEvidence.goalSlices`, not the generic purpose of the skill definition.
-- `skillDeclaredGoal` must describe the generic purpose declared by `skillContent`, not the concrete runtime user goal.
-- Keep `userGoal.slots`, `skillDeclaredGoal.keywords`, and `skillDeclaredGoal.expectedOutcomes` short keyword lists. Do not write long paragraphs there.
-- If `runtimeEvidence.userMessages` is empty or only contains protocol/runtime messages, set `userGoal.summary` to an empty string, `userGoal.slots` to `[]`, and runtime verdicts to `unknown`.
-- Use English only for enum values such as `passed`, `failed`, `router`, `frustrated`. All reviewer-facing text fields (`summary`, `slots`, `title`, `body`, `reviewerSummary`, `ownerSuggestions`, `acceptanceCriteria`) must be written in Chinese.
-- If evidence is insufficient, use `unknown`.
-- Keep output free of private user or session data beyond short evidence phrases already present in the input.
-- Parse each section independently. If one section is uncertain, still fill the other sections that can be judged.
-
-Output schema:
-
-```json
-{
-  "skillType": "router|delegation|executor|advisory|unknown",
-  "extractedStandards": {
-    "hardrules": [
-      {
-        "title": "Short reviewer-facing title",
-        "body": "Concrete rule a reviewer can check against evidence.",
-        "confidence": "low|medium|high",
-        "evidence": ["Short phrase from the skill definition or runtime summary"]
-      }
-    ],
-    "workflows": [
-      {
-        "title": "Workflow step title",
-        "body": "Concrete workflow behavior.",
-        "confidence": "low|medium|high",
-        "evidence": ["Short evidence phrase"]
-      }
-    ],
-    "completionCriteria": [
-      {
-        "title": "Completion criterion title",
-        "body": "How this skill should prove completion.",
-        "confidence": "low|medium|high",
-        "evidence": ["Short evidence phrase"]
-      }
-    ],
-    "artifactCriteria": [
-      {
-        "title": "Artifact criterion title",
-        "body": "What output artifact should exist and match.",
-        "confidence": "low|medium|high",
-        "evidence": ["Short evidence phrase"]
-      }
-    ]
-  },
-  "userGoal": {
-    "summary": "Short summary of the user's goal",
-    "slots": ["goal slot 1", "goal slot 2"],
-    "expectedOutcome": "Expected result or artifact"
-  },
-  "skillDeclaredGoal": {
-    "summary": "Short summary of what the skill claims to do",
-    "keywords": ["router", "consult", "PRD"],
-    "expectedOutcomes": ["child session", "artifact", "notification"]
-  },
-  "runtimeAssessment": {
-    "goalSatisfaction": "passed|failed|unknown",
-    "declaredBehaviorFit": "passed|failed|unknown",
-    "artifactGoalMatch": "passed|failed|unknown",
-    "userFeeling": "positive|neutral|negative|frustrated"
-  },
-  "userExperienceSignals": {
-    "useful": "passed|failed|unknown",
-    "followUp": "passed|failed|unknown",
-    "correction": "passed|failed|unknown",
-    "negativeFeedback": "passed|failed|unknown",
-    "interruption": "passed|failed|unknown",
-    "frustration": "passed|failed|unknown"
-  },
-  "reviewerSummary": "Evidence-backed summary that a reviewer can read quickly.",
-  "ownerSuggestions": [
-    {
-      "title": "Owner-facing fix title",
-      "body": "Concrete suggested fix.",
-      "evidence": ["Short evidence phrase"],
-      "acceptanceCriteria": "How the next review can tell this is fixed."
-    }
-  ]
-}
-```
-
-Allowed `skillType` values:
-- `router`
-- `delegation`
-- `executor`
-- `advisory`
-- `unknown`
-
-Allowed `confidence` values:
-- `low`
-- `medium`
-- `high`
-
-Judgment guidance:
-- Router skills are judged by route selection, goal preservation, downstream link, downstream completion, and user-facing closure.
-- Delegation skills are judged by parent/child contract, child lifecycle, parent boundary, output quality, and user notification.
-- Executor skills are judged by declared workflow execution, core tool fit, artifact creation, final delivery, and user feedback.
-- Advisory skills are judged by evidence quality, source traceability, conclusion coverage, and user feedback.