openhermes 4.3.0 → 4.11.2

package/harness/skills/oh-caveman/SKILL.md

@@ -1,42 +0,0 @@
----
-name: oh-caveman
-description: "Ultra-compressed communication mode — cut token usage ~75%"
-tier: 2
-triggers:
-  - "compress your response"
-  - "caveman mode"
-  - "shorter answers"
-route:
-  pass: mode
-  fail: mode
-  blocker: surface
----
-
-# oh-caveman
-
-## When to Use
-When context is tight, tokens are precious, or user says "caveman mode." Drops filler, articles, and pleasantries while keeping full technical accuracy.
-
-## Mode
-- No pleasantries, no hedging, no transitions
-- Fragments OK. One word when enough.
-- Short synonyms. Drop articles.
-- Code unchanged — only prose compresses.
-- Technical accuracy preserved at all costs.
-
-## Example
-Normal: "I think we should probably look at the authentication module because there might be an issue with the token refresh logic."
-Caveman: "Check auth module — token refresh likely broken."
-
-## Anti-patterns
-- Compressing code (code is already dense)
-- Omitting critical context to save tokens
-- Being unclear to be brief (accuracy > brevity)
-
-## Routing
-
-| Outcome | Route |
-|---------|-------|
-| pass | → [return to prior skill — mode active] |
-| fail | → [fallback to normal communication mode] |
-| blocker | → surface to user |

package/harness/skills/oh-learn/SKILL.md

@@ -1,101 +0,0 @@
----
-name: oh-learn
-description: "Extract, evolve, and promote session learnings as instincts. Review, search, prune, export."
-tier: 2
-triggers:
-  - "learn from session"
-  - "extract patterns"
-  - "run oh-learn"
-route:
-  pass: done
-  fail: surface
-  blocker: surface
----
-
-# oh-learn
-
-Learning engine for the harness. Distills patterns from sessions into **instincts** (trigger-action pairs with confidence), clusters them into skill candidates, and graduates high-signal patterns from project to global scope.
-
-## Instinct Data Model
-
-Every learning stored as one JSONL line in `~/.local/share/opencode/openhermes/plans/<project-name>-instincts.jsonl`:
-
-```json
-{ "trigger": "situation pattern", "action": "recommended response", "confidence": 0.5, "applications": 1, "successes": 1, "category": "coding", "source": "oh-learn:extract", "ts": "2026-05-15T12:00:00Z" }
-```
-
-**Rules:**
-- **Trigger** — specific, matchable situation. *Not* general advice.
-- **Action** — executable response. *Not* a belief.
-- **Confidence** — starts at 0.5, increments +0.05 per successful application, decays -0.02 per day without use.
-- **Category** — one of: `coding`, `testing`, `security`, `git`, `planning`, `orchestration`, `debugging`, `ux`.
-
-## When to Use
-
-After completing a significant piece of work, at session handoff, or when you notice the same pattern repeat 2+ times in one session. Also on explicit user request.
-
-## Workflows
-
-### Extract
-Mine the current session for reusable patterns.
-
-1. Scan recent conversation + code changes for repeated decision patterns
-2. For each distinct pattern write an instinct: trigger, action, confidence=0.5, category
-3. Read existing `~/.local/share/opencode/openhermes/plans/<project-name>-instincts.jsonl`, check for near-duplicate triggers
-4. If duplicate found: merge — `confidence = max(existing, 0.8 × new)`, increment applications
-5. If new: append line to file
-
-**Good instinct:** trigger=`"tsc --noEmit shows 10+ errors after batch edit"`, action=`"Fix errors one at a time, re-running tsc after each, rather than batch-fixing"`, category=`"debugging"`
-
-**Bad instinct:** `"Write clean code"` — too vague to trigger on.
-
-### Evolve
-Cluster related instincts into skill/command/agent candidates.
-
-1. Read all instincts from `~/.local/share/opencode/openhermes/plans/<project-name>-instincts.jsonl`
-2. Group by `category`, then by trigger topic similarity
-3. **If cluster ≥ 5 instincts AND avg confidence ≥ 0.7** → generate `oh-skill-craft` spec for a new skill
-4. **If cluster 3-4 instincts with confidence ≥ 0.8** → suggest update to existing skill
-5. Output candidate summary with trigger list and extracted core pattern
-
-### Promote
-Graduate high-confidence instincts from project to global scope.
-
-1. Scan `~/.local/share/opencode/openhermes/plans/<project-name>-instincts.jsonl` for instincts with `confidence >= 0.85 AND applications >= 10`
-2. Filter out project-specific patterns (reference paths, local APIs, domain terms)
-3. Append filtered candidates to `%USERPROFILE%\.config\opencode\instincts.jsonl` (global)
-4. Tag promoted instincts with `"promoted": true` in project file
-5. Report: "Promoted N instincts to global scope"
-
-### Review
-Show instinct summary: total count, confidence distribution, category breakdown, recently promoted.
-
-### Search
-Find instincts by topic, trigger fragment, category, or confidence range.
-
-### Prune
-Remove instincts stale for 30+ days with confidence < 0.3, or superseded by a higher-confidence instinct covering the same trigger.
-
-### Export
-Serialize instincts to portable JSON for sharing across projects or teams:
-
-```json
-{ "version": 1, "exported": "2026-05-15T12:00:00Z", "instincts": [...] }
-```
-
-## Anti-patterns
-
-- Hoarding every observation (most things aren't learnings)
-- Never pruning (stale knowledge is worse than no knowledge)
-- Storing what, not why (context-less facts are forgettable)
-- Over-promoting: not every pattern is globally useful
-- Extracting without applying: instincts that never trigger are noise
-- Ignoring confidence: treating all instincts as equally reliable
-
-## Routing
-
-| Outcome | Route |
-|---------|-------|
-| pass | → [done — report summary] |
-| fail | → [surface gaps to user] |
-| blocker | → surface to user |

package/lib/logger.ts

@@ -1,75 +0,0 @@
-import path from "node:path"
-import os from "node:os"
-import fs from "node:fs"
-
-export interface Logger {
-  debug: (...args: unknown[]) => void
-  info: (...args: unknown[]) => void
-  warn: (...args: unknown[]) => void
-  error: (...args: unknown[]) => void
-}
-
-const LEVELS: Record<string, number> = { debug: 0, info: 1, warn: 2, error: 3 }
-
-function resolveLevel(levelName: string | undefined): number | undefined {
-  if (!levelName) return undefined
-  return LEVELS[levelName as keyof typeof LEVELS]
-}
-
-const CURRENT_LEVEL = resolveLevel(process.env.OPENCODE_LOG_LEVEL?.trim().toLowerCase()) ?? (process.env.OPENHERMES_LOG_LEVEL?.trim().toLowerCase() === "debug" ? LEVELS.debug : LEVELS.warn)
-
-const LOG_DIR = path.join(os.homedir(), ".local", "share", "opencode", "log")
-const LOG_FILE = path.join(LOG_DIR, "openhermes.log")
-
-function ts(): string {
-  const d = new Date()
-  return `${d.getFullYear()}-${(d.getMonth()+1).toString().padStart(2,"0")}-${d.getDate().toString().padStart(2,"0")} ${d.getHours().toString().padStart(2,"0")}:${d.getMinutes().toString().padStart(2,"0")}:${d.getSeconds().toString().padStart(2,"0")}.${d.getMilliseconds().toString().padStart(3,"0")}`
-}
-
-function formatArgs(args: unknown[]): string {
-  return args.map(a => {
-    if (a === null) return "null"
-    if (a === undefined) return "undefined"
-    if (typeof a === "object") {
-      try { return (a as Error)?.message || JSON.stringify(a) } catch { return String(a) }
-    }
-    return String(a)
-  }).join(" ")
-}
-
-function shouldLog(levelName: string): boolean {
-  return LEVELS[levelName] >= CURRENT_LEVEL
-}
-
-let _fd: number | null = null
-function getFd(): number {
-  if (_fd) return _fd
-  try {
-    fs.mkdirSync(LOG_DIR, { recursive: true })
-    _fd = fs.openSync(LOG_FILE, "a")
-  } catch {
-    _fd = -1
-  }
-  return _fd
-}
-
-export function createLogger(name: string): Logger {
-  const prefix = `[openhermes:${name}]`
-
-  function emit(levelName: string, ...args: unknown[]): void {
-    if (!shouldLog(levelName)) return
-    const fd = getFd()
-    if (fd < 0) return
-    const line = `${ts()} ${prefix} [${levelName.toUpperCase()}] ${formatArgs(args)}\n`
-    try { fs.writeSync(fd, line) } catch {}
-  }
-
-  return {
-    debug: (...args: unknown[]) => emit("debug", ...args),
-    info: (...args: unknown[]) => emit("info", ...args),
-    warn: (...args: unknown[]) => emit("warn", ...args),
-    error: (...args: unknown[]) => emit("error", ...args),
-  }
-}
-
-export const rootLogger: Logger = createLogger("root")