openhermes 1.5.6 → 1.12.1

package/bootstrap.mjs

@@ -33,9 +33,9 @@ Snapshot before mutation. Never delete unrelated files. Never assume \`%USERPROF
 | Category | Items |
 |----------|-------|
 | **Native tools** | \`read\`, \`write\`, \`edit\`, \`glob\`, \`grep\`, \`bash\`, \`task\`, \`webfetch\`, \`skill\`, \`todowrite\`, \`todoread\` |
-| **In-process tools** | \`hm_put\`, \`hm_get\`, \`hm_list\`, \`hm_latest\`, \`hm_search\` |
+| **In-process tools** | \`add_memory\`, \`fetch_memory\`, \`list_memory\`, \`latest_memory\`, \`search_memory\`, \`archive_memory\` |
 | **Memory recall cache** | \`openhermes/memory/recall/cache.json\` — read on session start, no MCP round-trip |
-| **Subagents** | \`explore\` (read-only), \`general\` (multi-step), \`architect\`, \`planner\`, \`build-error-resolver\`, \`code-reviewer\`, \`security-reviewer\`, \`e2e-runner\` |
+| **Subagents** | \`explore\` (read-only), \`general\` (multi-step), \`architect\`, \`planner\`, \`build-error-resolver\`, \`code-reviewer\`, \`security-reviewer\`, \`e2e-runner\`, \`docs-lookup\`, \`doc-updater\`, \`refactor-cleaner\`, \`loop-operator\`, \`harness-optimizer\`, \`tdd-guide\`, \`review-go\`, \`build-go\`, \`review-database\`, \`review-cpp\`, \`build-cpp\`, \`review-java\`, \`build-java\`, \`review-kotlin\`, \`build-kotlin\`, \`review-python\`, \`review-rust\`, \`build-rust\` |
 | **Plugins** | \`curator\` (checkpoints, mistakes, audit, compaction), \`autorecall\` (recall cache on \`session.created\`), \`skill-builder\` (complex session detection) |
 
 ## Skills (available via \`skill\` tool)
@@ -56,17 +56,33 @@ Main context = coordination + verification only. Substantive work → subagent.
 | Security audit | \`security-reviewer\` |
 | E2E testing | \`e2e-runner\` |
 | Multi-file search/exploration | \`explore\` or \`general\` |
+| Documentation lookup | \`docs-lookup\` |
+| Doc/codemap update | \`doc-updater\` |
+| Dead code cleanup | \`refactor-cleaner\` |
+| TDD workflow | \`tdd-guide\` |
+| Autonomous loop | \`loop-operator\` |
+| Go review | \`review-go\` |
+| Go build fix | \`build-go\` |
+| Database review | \`review-database\` |
+| C++ review | \`review-cpp\` |
+| Java review | \`review-java\` |
+| Java build fix | \`build-java\` |
+| Kotlin review | \`review-kotlin\` |
+| Kotlin build fix | \`build-kotlin\` |
+| Python review | \`review-python\` |
+| Rust review | \`review-rust\` |
+| Rust build fix | \`build-rust\` |
 | Any non-trivial multi-step | appropriate specialist |
 
 Never delegate trivial single-step ops. Subagent returns diff + summary + verification; inspect return only. Full ref: \`${RULES_DIR}\\\\delegation.md\`.
 
 ## Memory — Gated & Precision-First
 
-- **Start**: Read recall cache first. If stale/missing → \`hm_latest\` for relevant classes.
-- **Before work**: Narrow \`hm_search\` by class, scope, keywords. Never read full indexes.
+- **Start**: Read recall cache first. If stale/missing → \`latest_memory\` for relevant classes.
+- **Before work**: Narrow \`search_memory\` by class, scope, keywords. Never read full indexes.
 - **Before close**: Query same-type mistakes (7 days). Match → \`code-reviewer\` or \`security-reviewer\`.
-- **On failure**: \`hm_search\` for similar incidents. Search memory before asking user.
-- **Precision ladder**: \`hm_latest\` → \`hm_search\` → \`hm_get\` → \`hm_list\` (last resort). Full index reads only for explicit audit/repair tasks.
+- **On failure**: \`search_memory\` for similar incidents. Search memory before asking user.
+- **Precision ladder**: \`latest_memory\` → \`search_memory\` → \`fetch_memory\` → \`list_memory\` (last resort). Full index reads only for explicit audit/repair tasks.
 - **Anti-spam**: No obvious facts, no one-off prefs, no temp state, no low-risk mistakes. Supersede, don't duplicate. Full rules: \`${RULES_DIR}\\\\retrieval.md\`, \`${RULES_DIR}\\\\memory-management.md\`.
 
 ## Self-Edit Authority
@@ -86,7 +102,7 @@ Full tiers: \`${RULES_DIR}\\\\self-heal.md\`.
 - Checkpoint on meaningful boundaries. Compress closed segments immediately.
 - After subagent return: verify → compress that block.
 - Compress proactively.
-- Skill candidates → \`/learn\` only if repeated pattern + \`hm_search\` confirms no dup. See \`${RULES_DIR}\\\\skills-management.md\`.
+- Skill candidates → \`/learn\` only if repeated pattern + \`search_memory\` confirms no dup. See \`${RULES_DIR}\\\\skills-management.md\`.
 - Audit triggers: openhermes/config change, repeated failures, session start when last audit >7 days. See \`${RULES_DIR}\\\\audit.md\`.
 
 ## Escalation
@@ -100,7 +116,7 @@ T0: observe → log mistake → smallest fix. T1: add prevention rule → verify
 - **Forensic ledger**: \`%USERPROFILE%\\\\.local\\\\share\\\\opencode\\\\opencode.db\``
 
   return [
-    `<OPENHERMES_BOOTSTRAP>\nOpenHermes v${getOwnVersion()} active. Harness: \`${HARNESS_DIR}\\\`. Memory at \`~/.config/opencode/openhermes/memory/\`. Rules at \`${RULES_DIR}\\\`. Skills discoverable via \`skill\` tool — use \`skill\` tool to list/load them.`,
+    `<OPENHERMES_BOOTSTRAP>\nOpenHermes v${getOwnVersion()} active. Harness: \`${HARNESS_DIR}\\\`. Memory at \`~/.local/share/opencode/openhermes/memory/\`. Rules at \`${RULES_DIR}\\\`. Skills discoverable via \`skill\` tool — use \`skill\` tool to list/load them.`,
     `<OPENHERMES_CONSTITUTION>\n${constitution}\n</OPENHERMES_CONSTITUTION>`,
     `<OPENHERMES_RUNTIME>\n${runtime}\n</OPENHERMES_RUNTIME>`,
     `<OPENHERMES_ROUTER>\n${router}\n</OPENHERMES_ROUTER>`
@@ -152,6 +168,32 @@ export const BootstrapPlugin = async ({ client, directory }) => {
         "doctor": { agent: "OpenHermes", description: "Run OpenCode OpenHermes health diagnostics", subtask: true, template: ct("doctor.md") },
         "memory-search": { agent: "OpenHermes", description: "Search OpenHermes memory with LLM summarization", subtask: true, template: ct("memory-search.md") },
         "learn": { agent: "OpenHermes", description: "Create a new skill from recent work patterns", subtask: true, template: ct("learn.md") },
+        "ohc": { template: "", description: "OHC context management: /ohc status, /ohc compress [focus]" },
+        "orchestrate": { agent: "planner", description: "Orchestrate multiple agents for complex tasks", subtask: true, template: ct("orchestrate.md") },
+        "eval": { agent: "planner", description: "Evaluate implementation against acceptance criteria", subtask: true, template: ct("eval.md") },
+        "model-route": { agent: "OpenHermes", description: "Recommend model tier by task complexity and budget", subtask: true, template: ct("model-route.md") },
+        "quality-gate": { agent: "OpenHermes", description: "Run quality pipeline (format, lint, type check)", subtask: true, template: ct("quality-gate.md") },
+        "test-coverage": { agent: "tdd-guide", description: "Analyze coverage reports and identify gaps", subtask: true, template: ct("test-coverage.md") },
+        "update-docs": { agent: "doc-updater", description: "Update documentation for recent code changes", subtask: true, template: ct("update-docs.md") },
+        "update-codemaps": { agent: "doc-updater", description: "Generate/update architecture codemaps", subtask: true, template: ct("update-codemaps.md") },
+        "refactor-clean": { agent: "refactor-cleaner", description: "Remove dead code and consolidate duplicates", subtask: true, template: ct("refactor-clean.md") },
+        "verify": { agent: "OpenHermes", description: "Run comprehensive verification loop (typecheck, lint, test, build)", subtask: true, template: ct("verify.md") },
+        "checkpoint": { agent: "OpenHermes", description: "Save verification state and progress checkpoint", subtask: true, template: ct("checkpoint.md") },
+        "loop-start": { agent: "loop-operator", description: "Start managed autonomous loop with safety defaults", subtask: true, template: ct("loop-start.md") },
+        "loop-status": { agent: "OpenHermes", description: "Inspect active loop state, progress, and failure signals", subtask: true, template: ct("loop-status.md") },
+        "harness-audit": { agent: "harness-optimizer", description: "Run harness self-audit across 7 categories", subtask: true, template: ct("harness-audit.md") },
+        "setup-pm": { agent: "OpenHermes", description: "Configure package manager preference for the project", subtask: true, template: ct("setup-pm.md") },
+        "go-build": { agent: "build-go", description: "Fix Go build, vet, and compilation errors", subtask: true, template: ct("go-build.md") },
+        "go-review": { agent: "review-go", description: "Review Go code for idiomatic patterns and best practices", subtask: true, template: ct("go-review.md") },
+        "rust-build": { agent: "build-rust", description: "Fix Rust build, clippy, and dependency errors", subtask: true, template: ct("rust-build.md") },
+        "rust-review": { agent: "review-rust", description: "Review Rust code for safety, ownership, and idioms", subtask: true, template: ct("rust-review.md") },
+        "skill-create": { agent: "OpenHermes", description: "Generate a new skill from git history analysis", subtask: true, template: ct("skill-create.md") },
+      }
+
+      config.experimental ??= {}
+      config.experimental.primary_tools ??= []
+      if (!config.experimental.primary_tools.includes("compress")) {
+        config.experimental.primary_tools.push("compress")
       }
 
       config.agent = {
@@ -210,6 +252,114 @@ export const BootstrapPlugin = async ({ client, directory }) => {
           prompt: p("security-reviewer.md"),
           permission: { read: "allow", edit: "deny", bash: "deny", task: { "*": "allow" } },
         },
+        "docs-lookup": {
+          description: "Documentation lookup via MCP — query any library docs in real-time",
+          mode: "subagent",
+          prompt: p("docs-lookup.md"),
+          permission: { read: "allow", bash: "allow", edit: "deny" },
+        },
+        "doc-updater": {
+          description: "Documentation and codemap generation/update specialist",
+          mode: "subagent",
+          prompt: p("doc-updater.md"),
+          permission: { read: "allow", edit: "allow", bash: "allow" },
+        },
+        "refactor-cleaner": {
+          description: "Dead code detection and safe removal specialist",
+          mode: "subagent",
+          prompt: p("refactor-cleaner.md"),
+          permission: { read: "allow", edit: "allow" },
+        },
+        "loop-operator": {
+          description: "Autonomous agent loop operator — safe iteration with stop conditions",
+          mode: "subagent",
+          prompt: p("loop-operator.md"),
+          permission: { read: "allow", edit: "allow", bash: "allow", task: { "*": "allow" } },
+        },
+        "harness-optimizer": {
+          description: "OpenHermes harness configuration optimizer — audit, tune, measure",
+          mode: "subagent",
+          prompt: p("harness-optimizer.md"),
+          permission: { read: "allow", bash: "allow", edit: "deny" },
+        },
+        "tdd-guide": {
+          description: "Test-Driven Development coach — red-green-refactor cycle enforcement",
+          mode: "subagent",
+          prompt: p("tdd-guide.md"),
+          permission: { read: "allow", edit: "allow", bash: "allow" },
+        },
+        "review-go": {
+          description: "Go code review specialist — idiomatic Go, concurrency, error handling",
+          mode: "subagent",
+          prompt: p("review-go.md"),
+          permission: { read: "allow", bash: "allow", edit: "deny" },
+        },
+        "build-go": {
+          description: "Go build error resolution specialist — go build, vet, staticcheck fixes",
+          mode: "subagent",
+          prompt: p("build-go.md"),
+          permission: { read: "allow", edit: "allow", bash: "allow" },
+        },
+        "review-database": {
+          description: "PostgreSQL database specialist — query optimization, schema, RLS, indexes",
+          mode: "subagent",
+          prompt: p("review-database.md"),
+          permission: { read: "allow", bash: "allow", edit: "deny" },
+        },
+        "review-cpp": {
+          description: "C++ code review specialist — memory safety, modern C++, RAII",
+          mode: "subagent",
+          prompt: p("review-cpp.md"),
+          permission: { read: "allow", bash: "allow", edit: "deny" },
+        },
+        "build-cpp": {
+          description: "C++ build error resolution specialist — CMake, linker, template errors",
+          mode: "subagent",
+          prompt: p("build-cpp.md"),
+          permission: { read: "allow", edit: "allow", bash: "allow" },
+        },
+        "review-java": {
+          description: "Java/Spring Boot review specialist — JPA, architecture, security",
+          mode: "subagent",
+          prompt: p("review-java.md"),
+          permission: { read: "allow", bash: "allow", edit: "deny" },
+        },
+        "build-java": {
+          description: "Java/Maven/Gradle build error resolution specialist",
+          mode: "subagent",
+          prompt: p("build-java.md"),
+          permission: { read: "allow", edit: "allow", bash: "allow" },
+        },
+        "review-kotlin": {
+          description: "Kotlin/Android review specialist — coroutines, Compose, architecture",
+          mode: "subagent",
+          prompt: p("review-kotlin.md"),
+          permission: { read: "allow", bash: "allow", edit: "deny" },
+        },
+        "build-kotlin": {
+          description: "Kotlin/Gradle build error resolution specialist",
+          mode: "subagent",
+          prompt: p("build-kotlin.md"),
+          permission: { read: "allow", edit: "allow", bash: "allow" },
+        },
+        "review-python": {
+          description: "Python code review specialist — PEP 8, type hints, security",
+          mode: "subagent",
+          prompt: p("review-python.md"),
+          permission: { read: "allow", bash: "allow", edit: "deny" },
+        },
+        "review-rust": {
+          description: "Rust code review specialist — ownership, lifetimes, safety",
+          mode: "subagent",
+          prompt: p("review-rust.md"),
+          permission: { read: "allow", bash: "allow", edit: "deny" },
+        },
+        "build-rust": {
+          description: "Rust build error resolution specialist — cargo, borrow checker, clippy",
+          mode: "subagent",
+          prompt: p("build-rust.md"),
+          permission: { read: "allow", edit: "allow", bash: "allow" },
+        },
       }
 
       config.default_agent = "OpenHermes"

package/curator.mjs

@@ -2,7 +2,7 @@ import path from "node:path"
 import fs from "node:fs"
 import os from "node:os"
 import { findUnsupportedSchemaKeywords, validateSchema } from "./lib/schema-validator.mjs"
-import { atomicWriteJson, fingerprintEnvironment, fingerprintFile, isTruthy, redactSensitiveText, sanitizeRecord, truncateText } from "./lib/hardening.mjs"
+import { atomicWriteJson, fingerprintEnvironment, fingerprintFile, readJson, redactSensitiveText, sanitizeRecord, truncateText } from "./lib/hardening.mjs"
 import { fileURLToPath } from "node:url"
 import { dirname } from "node:path"
 import { getDataRoot, getMemoryRoot, getRuntimeRoot, getArchiveRoot } from "./lib/paths.mjs"
@@ -20,10 +20,6 @@ function curatorLog(message) {
   process.stderr.write(`${message}\n`)
 }
 
-function readJson(fp, fallback) {
-  try { return JSON.parse(fs.readFileSync(fp, "utf8")) } catch { return fallback }
-}
-
 function buildEnvironmentFingerprint(root, directory, project) {
   return fingerprintEnvironment({
     cwd: directory,

package/harness/commands/checkpoint.md

@@ -0,0 +1,68 @@
+---
+description: Save verification state and progress checkpoint
+agent: OpenHermes
+subtask: true
+---
+
+# Checkpoint Command
+
+Save current verification state and create progress checkpoint: $ARGUMENTS
+
+## Your Task
+
+Create a snapshot of current progress including:
+
+1. **Tests status** - Which tests pass/fail
+2. **Coverage** - Current coverage metrics
+3. **Build status** - Build succeeds or errors
+4. **Code changes** - Summary of modifications
+5. **Next steps** - What remains to be done
+
+## Checkpoint Format
+
+### Checkpoint: [Timestamp]
+
+**Tests**
+- Total: X
+- Passing: Y
+- Failing: Z
+- Coverage: XX%
+
+**Build**
+- Status: PASS: Passing / FAIL: Failing
+- Errors: [if any]
+
+**Changes Since Last Checkpoint**
+```
+git diff --stat [last-checkpoint-commit]
+```
+
+**Completed Tasks**
+- [x] Task 1
+- [x] Task 2
+- [ ] Task 3 (in progress)
+
+**Blocking Issues**
+- [Issue description]
+
+**Next Steps**
+1. Step 1
+2. Step 2
+
+## Usage with Verification Loop
+
+Checkpoints integrate with the verification loop:
+
+```
+/plan → implement → /checkpoint → /verify → /checkpoint → implement → ...
+```
+
+Use checkpoints to:
+- Save state before risky changes
+- Track progress through phases
+- Enable rollback if needed
+- Document verification points
+
+---
+
+**TIP**: Create checkpoints at natural breakpoints: after each phase, before major refactoring, after fixing critical bugs.

package/harness/commands/eval.md

@@ -0,0 +1,89 @@
+---
+description: Run evaluation against acceptance criteria
+agent: planner
+subtask: true
+---
+
+# Eval Command
+
+Evaluate implementation against acceptance criteria: $ARGUMENTS
+
+## Your Task
+
+Run structured evaluation to verify the implementation meets requirements.
+
+## Evaluation Framework
+
+### Grader Types
+
+1. **Binary Grader** - Pass/Fail
+   - Does it work? Yes/No
+   - Good for: feature completion, bug fixes
+
+2. **Scalar Grader** - Score 0-100
+   - How well does it work?
+   - Good for: performance, quality metrics
+
+3. **Rubric Grader** - Category scores
+   - Multiple dimensions evaluated
+   - Good for: comprehensive review
+
+## Evaluation Process
+
+### Step 1: Define Criteria
+
+```
+Acceptance Criteria:
+1. [Criterion 1] - [weight]
+2. [Criterion 2] - [weight]
+3. [Criterion 3] - [weight]
+```
+
+### Step 2: Run Tests
+
+For each criterion:
+- Execute relevant test
+- Collect evidence
+- Score result
+
+### Step 3: Calculate Score
+
+```
+Final Score = Σ (criterion_score × weight) / total_weight
+```
+
+### Step 4: Report
+
+## Evaluation Report
+
+### Overall: [PASS/FAIL] (Score: X/100)
+
+### Criterion Breakdown
+
+| Criterion | Score | Weight | Weighted |
+|-----------|-------|--------|----------|
+| [Criterion 1] | X/10 | 30% | X |
+| [Criterion 2] | X/10 | 40% | X |
+| [Criterion 3] | X/10 | 30% | X |
+
+### Evidence
+
+**Criterion 1: [Name]**
+- Test: [what was tested]
+- Result: [outcome]
+- Evidence: [screenshot, log, output]
+
+### Recommendations
+
+[If not passing, what needs to change]
+
+## Pass@K Metrics
+
+For non-deterministic evaluations:
+- Run K times
+- Calculate pass rate
+- Report: "Pass@K = X/K"
+
+---
+
+**TIP**: Use eval for acceptance testing before marking features complete.

package/harness/commands/go-build.md

@@ -0,0 +1,87 @@
+---
+description: Fix Go build and vet errors
+agent: build-go
+subtask: true
+---
+
+# Go Build Command
+
+Fix Go build, vet, and compilation errors: $ARGUMENTS
+
+## Your Task
+
+1. **Run go build**: `go build ./...`
+2. **Run go vet**: `go vet ./...`
+3. **Fix errors** one by one
+4. **Verify fixes** don't introduce new errors
+
+## Common Go Errors
+
+### Import Errors
+```
+imported and not used: "package"
+```
+**Fix**: Remove unused import or use `_` prefix
+
+### Type Errors
+```
+cannot use x (type T) as type U
+```
+**Fix**: Add type conversion or fix type definition
+
+### Undefined Errors
+```
+undefined: identifier
+```
+**Fix**: Import package, define variable, or fix typo
+
+### Vet Errors
+```
+printf: call has arguments but no formatting directives
+```
+**Fix**: Add format directive or remove arguments
+
+## Fix Order
+
+1. **Import errors** - Fix or remove imports
+2. **Type definitions** - Ensure types exist
+3. **Function signatures** - Match parameters
+4. **Vet warnings** - Address static analysis
+
+## Build Commands
+
+```bash
+# Build all packages
+go build ./...
+
+# Build with race detector
+go build -race ./...
+
+# Build for specific OS/arch
+GOOS=linux GOARCH=amd64 go build ./...
+
+# Run go vet
+go vet ./...
+
+# Run staticcheck
+staticcheck ./...
+
+# Format code
+gofmt -w .
+
+# Tidy dependencies
+go mod tidy
+```
+
+## Verification
+
+After fixes:
+```bash
+go build ./...    # Should succeed
+go vet ./...      # Should have no warnings
+go test ./...     # Tests should pass
+```
+
+---
+
+**IMPORTANT**: Fix errors only. No refactoring, no improvements. Get the build green with minimal changes.

package/harness/commands/go-review.md

@@ -0,0 +1,71 @@
+---
+description: Go code review for idiomatic patterns
+agent: review-go
+subtask: true
+---
+
+# Go Review Command
+
+Review Go code for idiomatic patterns and best practices: $ARGUMENTS
+
+## Your Task
+
+1. **Analyze Go code** for idioms and patterns
+2. **Check concurrency** - goroutines, channels, mutexes
+3. **Review error handling** - proper error wrapping
+4. **Verify performance** - allocations, bottlenecks
+
+## Review Checklist
+
+### Idiomatic Go
+- [ ] Package naming (lowercase, no underscores)
+- [ ] Variable naming (camelCase, short)
+- [ ] Interface naming (ends with -er)
+- [ ] Error naming (starts with Err)
+
+### Error Handling
+- [ ] Errors are checked, not ignored
+- [ ] Errors wrapped with context (`fmt.Errorf("...: %w", err)`)
+- [ ] Sentinel errors used appropriately
+- [ ] Custom error types when needed
+
+### Concurrency
+- [ ] Goroutines properly managed
+- [ ] Channels buffered appropriately
+- [ ] No data races (use `-race` flag)
+- [ ] Context passed for cancellation
+- [ ] WaitGroups used correctly
+
+### Performance
+- [ ] Avoid unnecessary allocations
+- [ ] Use `sync.Pool` for frequent allocations
+- [ ] Prefer value receivers for small structs
+- [ ] Buffer I/O operations
+
+### Code Organization
+- [ ] Small, focused packages
+- [ ] Clear dependency direction
+- [ ] Internal packages for private code
+- [ ] Godoc comments on exports
+
+## Report Format
+
+### Idiomatic Issues
+- [file:line] Issue description
+  Suggestion: How to fix
+
+### Error Handling Issues
+- [file:line] Issue description
+  Suggestion: How to fix
+
+### Concurrency Issues
+- [file:line] Issue description
+  Suggestion: How to fix
+
+### Performance Issues
+- [file:line] Issue description
+  Suggestion: How to fix
+
+---
+
+**TIP**: Run `go vet` and `staticcheck` for additional automated checks.

package/harness/commands/harness-audit.md

@@ -0,0 +1,90 @@
+---
+description: Run a self-evaluation audit of the OpenHermes harness
+agent: harness-optimizer
+subtask: true
+---
+
+# Harness Audit Command
+
+Run a self-evaluation audit of the OpenHermes harness and return a prioritized scorecard.
+
+## Usage
+
+`/harness-audit [scope] [--format text|json] [--root path]`
+
+- `scope` (optional): `repo` (default), `hooks`, `skills`, `commands`, `agents`
+- `--format`: output style (`text` default, `json` for automation)
+- `--root`: audit a specific path instead of the current working directory
+
+## Self-Evaluation Checklist
+
+Evaluate each category by inspecting the harness files directly. No external script needed.
+
+### 1. Tool Coverage (0-10)
+- [ ] Commands exist for all subagent types
+- [ ] Each command has correct frontmatter (description, agent, subtask)
+- [ ] Agent mapping table is complete and accurate
+- [ ] Language-specific agents exist (Go, Rust)
+- [ ] All command files are discoverable
+
+### 2. Context Efficiency (0-10)
+- [ ] Commands are concise (under 100 lines each)
+- [ ] No redundant or overlapping command content
+- [ ] Delegation instructions are clear
+- [ ] Subagent handoffs minimize context overhead
+- [ ] Templates provide structured output formats
+
+### 3. Quality Gates (0-10)
+- [ ] verify.md has comprehensive checklist
+- [ ] quality-gate.md covers lint/type/build
+- [ ] test-coverage.md has meaningful targets
+- [ ] eval.md has structured scoring framework
+- [ ] checkpoint.md enables state tracking
+
+### 4. Memory Persistence (0-10)
+- [ ] Memory tools documented (add_memory/fetch_memory/list_memory/latest_memory/search_memory)
+- [ ] Checkpoint command references memory persistence
+- [ ] Mistake/audit logging workflow documented
+- [ ] Recall cache strategy defined
+
+### 5. Eval Coverage (0-10)
+- [ ] eval.md supports binary/scalar/rubric grading
+- [ ] Acceptance criteria framework in place
+- [ ] Pass@K metrics for non-deterministic evals
+- [ ] Evaluation report format defined
+
+### 6. Security Guardrails (0-10)
+- [ ] verify.md includes security checklist items
+- [ ] No hardcoded secrets guidance present
+- [ ] Input validation guidance included
+- [ ] SQL injection / XSS risks addressed
+
+### 7. Cost Efficiency (0-10)
+- [ ] model-route.md has budget tiers
+- [ ] Model routing heuristic defined
+- [ ] Subagent usage reduces main-context tokens
+- [ ] Compression workflow documented
+
+## Output Contract
+
+Return:
+
+1. `overall_score` out of `max_score` (70 for `repo`; smaller for scoped audits)
+2. Category scores and concrete findings
+3. Failed checks with exact file paths
+4. Top 3 actions to improve
+5. Suggested commands to apply next
+
+## Example Result
+
+```text
+Harness Audit (repo): 66/70
+- Tool Coverage: 10/10 (10/10 pts)
+- Context Efficiency: 9/10 (9/10 pts)
+- Quality Gates: 10/10 (10/10 pts)
+
+Top 3 Actions:
+1) [Security Guardrails] Add preflight security guard instructions in verify.md.
+2) [Tool Coverage] Ensure all subagent types have corresponding command files.
+3) [Eval Coverage] Add evaluation templates and examples.
+```

package/harness/commands/learn.md

@@ -11,7 +11,7 @@ Create a new reusable skill from recent work patterns. $ARGUMENTS
 ## Your Task
 
 1. **Search backlog** for pending skill candidates:
-   - Use `hm_search` with query="skill-candidate" classes=["backlog"]
+   - Use `search_memory` with query="skill-candidate" classes=["backlog"]
    - If $ARGUMENTS is non-empty, narrow search to that topic
 2. **Analyze the candidate** — what pattern did the session reveal?
 3. **Create the skill**:
@@ -19,7 +19,7 @@ Create a new reusable skill from recent work patterns. $ARGUMENTS
    - Follow its instructions to create a new SKILL.md
    - Target: `%USERPROFILE%\.config\opencode\skills\<name>\SKILL.md`
    - Naming: lowercase, hyphenated, descriptive
-4. **Close the backlog entry**: `hm_put(class="backlog", id="<candidate-id>", data={..., status:"closed"})`
+4. **Close the backlog entry**: `add_memory(class="backlog", id="<candidate-id>", data={..., status:"closed"})`
 5. **Report**: What skill was created, where, and what it does
 
 ## Skill Requirements