oh-my-agent-commandcode 0.0.0-stage → 20.38.3

package/skills/oma-scm/SKILL.md

@@ -0,0 +1,103 @@
+---
+name: oma-scm
+description: Manage Git branches, merges, conflicts, commits, and release
+  baselines. Use for repository history and change-management operations.
+---
+
+# Software Configuration Management
+
+## Scheduling
+
+### Goal
+Perform Git operations with explicit scope, traceable commits, and recoverable history.
+
+### Intent signature
+- Commit, stage, push, branch, merge, rebase, resolve conflicts, manage worktrees, or inspect SCM governance.
+
+### When to use
+- Commit and push requests, including Conventional Commit messages and logical splitting.
+- Branch/history operations, releases, baselines, CODEOWNERS, and configuration-management reviews.
+
+### When NOT to use
+- Implementing a feature or fixing application code -> use the owning specialist.
+- General requirements planning -> use `oma-pm`; security/testing review -> use `oma-qa`.
+
+### Expected inputs
+Requested Git operation, repository state, and effective `scm` settings from project configuration.
+
+### Expected outputs
+Requested Git changes or an advisory plan, commit/branch identifiers, checks performed, and unresolved work.
+
+### Dependencies
+Git CLI; project configuration and hooks. Read the references below only for the selected operation.
+
+## Structural Flow
+
+### Entry
+Inspect branch, upstream, staged/unstaged changes, and existing authorization. Select the commit path below or the configuration-management resource for broader operations.
+
+### Transitions
+- Commit-only request: finish after committing. Push or create a PR only when requested or required by the applicable workflow.
+- Independent changes: split by feature. One logical change remains one commit even across code, tests, and docs; an explicit grouping instruction wins. File count is only a tiebreaker (at most five files lean single).
+- Governance, worktrees, releases, or history operations: load `resources/cm-operations.md`.
+- Large merges: load `resources/merge-risk.md` before selecting merge order or recovery steps.
+- Default-branch push: apply the Push and PR safety section below.
+
+### Failure and recovery
+| Failure | Recovery |
+|---|---|
+| Unrelated dirty or staged files | Keep them outside the requested commit; never silently absorb them into an amend |
+| Commit hook rejects a message | Fix the message or actual defect according to the hook; do not bypass it |
+| Push is non-fast-forward | Fetch, inspect divergence, and integrate locally; do not retry with force |
+| Conflicting changes | Preserve both intents, resolve, and run affected checks before completion |
+| Likely secrets in the proposed diff | Stop before staging and identify the affected path without exposing the value |
+
+### Exit
+Report created commits/refs and verification. A failed commit, rejected push, or unresolved conflict is not completion.
+
+## Logical Operations
+
+### Canonical command path
+1. Inspect the worktree and recent conventions:
+   ```bash
+   git status -sb
+   git diff --staged
+   git diff
+   git log --oneline -5
+   ```
+2. Select logical commit groups under Transitions. Read `resources/conventional-commits.md` for message syntax, type, footer, and branch naming; apply repository hook/config limits.
+3. Show the selected message. Stage explicit paths, inspect the staged diff, and commit using the prepared message file:
+   ```bash
+   git add -- <specific-files>
+   git diff --cached --check
+   git diff --cached
+   git commit -F <message-file>
+   ```
+4. Inspect the resulting commit and remaining worktree. If push was requested, follow Push and PR safety, then report the remote result.
+
+### Guardrails
+1. Explicit, unambiguous user instructions take precedence over SCM defaults, including commit grouping and direct default-branch pushes. Existing authorization persists; do not ask for it again. Likely-secret material still requires stopping before staging.
+2. Stage explicit paths. Do not use `git add -A` or `git add .` without explicit authorization, and never include credentials or secret files.
+3. Do not rewrite shared history without explicit authorization. For an authorized rewrite, use `--force-with-lease`, never plain `--force`.
+4. Read the staged diff before an amend. Unpushed commits may be amended or reorganized within the requested scope; pushed commits are shared history.
+5. User-facing responses follow the configured language; commit messages, branch names, and PR titles/bodies stay in English.
+
+### Push and PR safety (only when requested)
+- Confirm the branch, upstream, and ahead/behind state before pushing.
+- If `scm.require_pr_for_default_branch` is true, use a topic branch and PR for default-branch changes, unless the user explicitly requested a direct push.
+- Keep the repository's required hooks/checks. A failure requires repair or an accurate partial result, not an assertion of success.
+- Verify that the requested commits reached the intended remote branch.
+
+### Amend, fixup, autosquash
+Determine whether the target commit is shared using `git log --oneline @{u}..HEAD` when an upstream exists. If there is no upstream, inspect remote refs before assuming a commit is unshared. Keep unrelated staged changes out of the operation.
+
+### Resource scope and effects
+Git operations change the index, local commits/refs, worktrees, and possibly remote refs. Read only relevant project configuration; never copy credential values into messages or reports. Apply `../_shared/core/execution-policy.md` to authorization and verification.
+
+## References
+- Commit syntax, types, co-author policy, and branch naming: `resources/conventional-commits.md` (commit requests)
+- Configuration-management operations: `resources/cm-operations.md` (governance/history/worktree tasks)
+- Merge risk and rollback: `resources/merge-risk.md` (large merges)
+- Ownership detail: `resources/codeowners-playbook.md` (CODEOWNERS work)
+- Onboarding indicators: `resources/onboarding-risk-signals.md` (repository-risk assessment)
+- Release observability: `../oma-observability/SKILL.md` §Integrations (release markers and baseline comparisons)

package/skills/oma-scm/resources/cm-operations.md

@@ -0,0 +1,56 @@
+# SCM Operations
+
+## 1) Planning
+
+1. Read the effective `scm` configuration and files listed under `documented_process`.
+2. If missing, infer from `CONTRIBUTING.md` / `README`; state assumptions.
+3. Confirm **branching model** and whether **force-push** on shared branches is allowed (default: not without explicit approval).
+
+## 2) Identification
+
+1. Canonical refs: default branch, release branches/tags, version sources (`package.json`, etc.).
+2. `.gitattributes` / LFS for binaries and generated assets.
+3. Branch names vs the effective `scm` configuration `branch_prefixes` when the project uses them.
+
+## 3) Control
+
+1. Small, reviewable units; align commits with PR / issue intent.
+2. **Conflicts:** `merge-base`, `git status`, resolve markers, tests; suggest `rerere` when conflicts repeat.
+3. **Worktrees:** `git worktree add`; merge/rebase from the **target branch’s** checkout; all worktrees share one object database.
+4. Do not rewrite **shared** history without maintainer approval; prefer `--force-with-lease` if force-push is unavoidable.
+
+## 4) Status accounting
+
+1. `git status -sb`: branch, remote tracking, ahead/behind, merge state.
+2. Relate last tag / release branch to `CHANGELOG` or tooling (semantic-release, release-please, changesets) if present.
+
+## 5) Verification & audit
+
+1. Required CI and `merge_group` when merge queue applies.
+2. Never stage/commit secrets (`.env`, keys, raw tokens). Filename patterns from the effective `scm` configuration `forbidden_patterns` are enforced mechanically by the `scm-guard` PreToolUse hook; for content-level leaks (tokens hardcoded in ordinary source files), run a scanner when available (`gitleaks protect --staged`, `trufflehog git`) before large or unfamiliar commits.
+3. Call out signed-commit expectations when the org cares about verification badges.
+
+### CODEOWNERS maintenance checklist
+
+1. Validate CODEOWNERS file exists (prefer `.github/CODEOWNERS`).
+2. Ensure critical paths are explicitly owned (not only fallback `*`).
+3. Ensure owners are active and mapped to current teams.
+4. Confirm branch protection requires CODEOWNERS review where needed.
+5. Flag overlapping/ambiguous rules that can hide intended owners.
+
+Read `change_governance.require_codeowners` and `ownership.*` in the effective `scm` configuration when present.
+
+## 6) Onboarding risk scan (optional)
+
+Use this quick scan when joining or inheriting a repository to identify risky areas before major changes.
+
+1. High churn files in `lookback` window.
+2. Ownership concentration / bus-factor signals.
+3. Bug hotspot files from fix-related history.
+4. Velocity trend by month.
+5. Revert/hotfix/emergency frequency.
+
+Read thresholds from the effective `scm` configuration `onboarding_metrics` when present and cite caveats:
+- squash merge teams can distort ownership metrics,
+- weak commit labeling reduces hotspot accuracy,
+- monorepo commit counts can bias subsystem interpretation.

package/skills/oma-scm/resources/codeowners-playbook.md

@@ -0,0 +1,45 @@
+# CODEOWNERS Playbook (SCM)
+
+Use this checklist to keep ownership and approval paths healthy.
+
+## What to check
+
+1. `CODEOWNERS` exists in one supported location:
+   - `.github/CODEOWNERS`
+   - `CODEOWNERS` (repo root)
+   - `docs/CODEOWNERS`
+2. Critical paths are explicitly owned (avoid relying only on `*` wildcard).
+3. Owners in rules are active maintainers (not departed/inactive accounts).
+4. Branch protection requires CODEOWNERS review for protected branches.
+5. Rules are minimally overlapping and ordered intentionally.
+
+## Baseline template
+
+```text
+# Global fallback
+* @org/maintainers
+
+# CI and release control
+.github/workflows/* @org/devops
+cli/commands/migrations/* @org/release-engineering
+
+# Agent system and governance
+.agents/hooks/** @org/agent-platform
+.agents/workflows/** @org/agent-platform
+
+# Product docs/content
+web/content/** @org/docs
+docs/** @org/docs
+```
+
+## Common anti-patterns
+
+- Only global `*` rule with no critical-path overrides.
+- Owners set to individuals only (no team alias).
+- Stale owners after team changes.
+- Critical infra/security paths missing explicit owners.
+
+## Review cadence
+
+- Re-check CODEOWNERS at least once per quarter.
+- Re-check immediately after org/team restructure.

package/skills/oma-scm/resources/conventional-commits.md

@@ -0,0 +1,108 @@
+# Conventional Commits Guide
+
+Git execution and staging rules live in the parent SKILL.md. Use this reference only for commit syntax and branch names. Repository configuration and hooks determine applicable length and release rules.
+
+## Commit Message Structure
+
+```
+<type>[optional scope]: <description>
+
+[optional body]
+
+[optional footer(s)]
+```
+
+## Types
+
+### Primary Types
+
+| Type | Description | SemVer | Example |
+|------|-------------|--------|---------|
+| `feat` | Add new feature | MINOR | `feat: add user authentication` |
+| `fix` | Bug fix | PATCH | `fix: resolve login timeout issue` |
+
+### Secondary Types
+
+| Type | Description | SemVer | Example |
+|------|-------------|--------|---------|
+| `docs` | Documentation changes | - | `docs: update API documentation` |
+| `style` | Code style changes (formatting, semicolons, etc.) | - | `style: fix indentation` |
+| `refactor` | Code improvement without behavior change | - | `refactor: extract helper function` |
+| `perf` | Performance improvements | PATCH | `perf: optimize database queries` |
+| `test` | Add/modify tests | - | `test: add unit tests for auth` |
+| `chore` | Maintenance not covered by other types | - | `chore: update .gitignore` |
+| `build` | Build system, external dependencies | - | `build: bump vite to v6` |
+| `ci` | CI configuration and scripts | - | `ci: cache bun install in workflow` |
+| `revert` | Revert a previous commit | varies | `revert: feat(auth): add OAuth2 support` |
+
+## Scope
+
+Scope indicates the area of changed code:
+
+```
+feat(auth): add OAuth2 support
+fix(api): handle null response
+refactor(ui): simplify button component
+```
+
+## Description
+
+- **Imperative mood**: "add", "fix", "update" (NOT "added", "fixed", "updates")
+- **Lowercase first letter**
+- **No trailing period**
+- **72 characters or less**
+
+## Body
+
+Body is optional but useful for complex changes:
+
+```
+feat(auth): add multi-factor authentication
+
+Implement TOTP-based two-factor authentication:
+- Add QR code generation for authenticator apps
+- Store encrypted TOTP secrets in database
+- Add backup codes for account recovery
+
+Closes #123
+```
+
+## Breaking Changes
+
+Breaking changes marked with `!` or in footer:
+
+```
+feat(api)!: change response format for user endpoint
+
+BREAKING CHANGE: The user endpoint now returns a nested object
+instead of a flat structure. Update client code accordingly.
+```
+
+## Footer
+
+### Issue References
+```
+feat(auth): add password reset flow
+
+Closes #456
+Refs #123, #789
+```
+
+### Co-Authors
+
+Include `Co-authored-by: <name> <email>` only when effective `scm.co_author.enabled` is true and both configured values are present. Copy them exactly from configuration; otherwise omit the trailer. GitHub credits the verified email owner, so never reuse an example identity. A co-author hook rejection must be corrected against the configured allowlist.
+
+## Branch Naming Convention
+
+| Type | Branch Prefix | Example |
+|------|---------------|---------|
+| feat | `feature/` | `feature/user-auth` |
+| fix | `fix/` | `fix/login-timeout` |
+| refactor | `refactor/` | `refactor/api-cleanup` |
+| docs | `docs/` | `docs/api-guide` |
+| hotfix | `hotfix/` | `hotfix/security-patch` |
+
+## Resources
+
+- [Conventional Commits Specification](https://www.conventionalcommits.org/)
+- [Semantic Versioning](https://semver.org/)

package/skills/oma-scm/resources/merge-risk.md

@@ -0,0 +1,76 @@
+# Large Merge Risk Triage
+
+Trigger this step when merge scope is large by change footprint, not PR count.
+Read thresholds from `.agents/oma-config.yaml` `large_merge_thresholds.*` first.
+If config values are missing, use these defaults:
+- combined changed files >= 150
+- combined additions+deletions >= 3000 lines
+- touching >= 3 high-churn/hotspot paths
+- any candidate has `risk_score >= 60`
+
+Use these signals:
+- file overlap across PRs (same files)
+- line-range overlap when available
+- branch age and divergence from base
+- hotspot files (high churn/recent edits)
+- ownership spread (many authors/teams touching same area)
+- semantic flags (API contract/interface/schema changes)
+
+Risk score formula (0-100):
+
+`risk_score = overlap(0-40) + divergence(0-20) + hotspot(0-15) + ownership(0-15) + semantic(0-10)`
+
+Bucket thresholds:
+- **LOW** (0-29): no overlap, low divergence, no semantic flags
+- **MEDIUM** (30-59): partial overlap or moderate divergence
+- **HIGH** (60-100): line overlap, repeated hotspot collisions, or semantic flags
+
+Scoring guidance:
+- `overlap`: 0 (none), 20 (same file only), 40 (same file + overlapping lines)
+- `divergence`: 0 (<24h and <=10 commits behind), 10 (1-3 days or <=50 behind), 20 (>3 days or >50 behind)
+- `hotspot`: 0 (stable), 8 (moderate churn), 15 (top churn paths touched)
+- `ownership`: 0 (single owner/team), 8 (2-3 owners), 15 (cross-team and unclear ownership)
+- `semantic`: 0 (none), 5 (minor contract touch), 10 (API/schema/interface breaking risk)
+
+Data sources (preferred order):
+1. PR metadata/diff from GitHub CLI or API
+2. Line-overlap detection — compare diff hunk ranges across candidate branches (no dedicated tool required):
+   ```bash
+   # Resolve ONE common base for both candidate branches.
+   COMMON_BASE=$(git merge-base <branch-a> <branch-b>)
+   git diff -U0 "$COMMON_BASE" <branch-a> -- <file> | grep '^@@'
+   git diff -U0 "$COMMON_BASE" <branch-b> -- <file> | grep '^@@'
+   # Compare the "-start,count" ranges in this common original file.
+   # A missing count means 1. Pure insertions (count=0) use the boundary
+   # after start; compare insertion boundaries and edits touching them.
+   # Do not compare "+start,count": each branch has different new-line offsets.
+   ```
+3. Merge simulation (GitHub mergeability/queue simulation when available)
+4. Local git history for churn/hotspot and ownership hints
+
+For large-scope merges, propose merge order as:
+1. LOW in small batches
+2. MEDIUM in smaller batches
+3. HIGH one-by-one with explicit checkpoints
+
+## Authorization for risky operations
+
+**Precedence:** an explicit, unambiguous user instruction overrides this gate (same as `oma-scm` SKILL.md Guardrails) — if the user already told you exactly what to do, state what you are doing and proceed without re-confirming. The single exception that always warrants a heads-up is likely-secret material. The gate below applies when the risky condition was NOT explicitly requested by the user.
+
+Stop and ask user confirmation if any of these are true:
+- merge conflicts are already present
+- history rewrite is required (`--force`, `reset --hard`, destructive restore/clean)
+- required checks, required reviews, or CODEOWNERS conditions are not satisfied
+- protected/main branch policy could be violated
+- release-critical paths are involved and rollback plan is unclear
+
+Additional confirmation triggers:
+- `risk_score >= 60`
+- batch failure repeated 2+ times
+- merge queue is unavailable and manual direct-merge is requested
+
+## Failure and rollback
+- On batch failure, bisect once and retry with smaller batch.
+- If the second attempt fails, stop and apply the authorization rules above.
+- Never continue high-risk merges after repeated failures without explicit approval.
+- For protected/main branches, prefer revert-based rollback over history rewrite.

package/skills/oma-scm/resources/onboarding-risk-signals.md

@@ -0,0 +1,71 @@
+# Onboarding Risk Signals (SCM)
+
+Use these lightweight SCM metrics to quickly estimate maintenance risk in an unfamiliar repository.
+
+## 1) High-Churn Files
+
+Frequent edits often indicate maintenance hotspots or unstable boundaries.
+
+```bash
+git log --format=format: --name-only --since="1 year ago" | sort | uniq -c | sort -nr | head -20
+```
+
+## 2) Ownership / Bus Factor
+
+Detect concentration risk and potential knowledge silos.
+
+```bash
+git shortlog -sn --no-merges HEAD
+```
+
+Heuristic:
+- One author > 60% of commits can indicate bus-factor risk.
+- If top contributors are inactive for 6+ months, flag continuity risk.
+- With squash-merge-heavy teams, treat commit-count ownership as approximate.
+
+## 3) Bug Hotspots
+
+Approximate bug-prone files from bug-fix commit history.
+
+```bash
+git log -i -E --grep="fix|bug|broken" --name-only --format='' | sort | uniq -c | sort -nr | head -20
+```
+
+Cross-reference with high-churn list to identify high-risk files.
+
+## 4) Development Velocity Trend
+
+Inspect monthly activity and momentum shifts.
+
+```bash
+git log --format='%ad' --date=format:'%Y-%m' | sort | uniq -c
+```
+
+Heuristic:
+- Stable/increasing trend suggests healthy delivery cadence.
+- 50%+ sudden drop can indicate staffing or ownership disruption.
+- 6-12 month steady decline can indicate project stagnation.
+
+## 5) Revert / Hotfix / Firefighting Signal
+
+Estimate operational reliability from emergency recovery patterns.
+
+```bash
+git log --oneline --since="1 year ago" | rg -i "revert|hotfix|emergency|rollback"
+```
+
+Heuristic:
+- A few incidents per year can be normal.
+- Incidents every ~2 weeks suggest unstable release/testing process.
+- Zero incidents may mean stable delivery or poor commit labeling.
+
+## Practical Output Template
+
+Report these sections:
+- Top churn files (top 10-20)
+- Ownership concentration and inactive key contributors
+- Bug hotspot overlaps with churn files
+- Velocity trend summary (recent 6-12 months)
+- Firefighting frequency and confidence level
+
+Always include caveats: squash merges, inconsistent commit labels, and monorepo skew.

package/skills/oma-search/SKILL.md

@@ -0,0 +1,125 @@
+---
+name: oma-search
+description: Find external documentation, web sources, or remote code with
+  citations. Local code navigation uses the configured code tools.
+---
+
+# Search Router
+
+## Scheduling
+
+### Goal
+Select the search or retrieval channel needed by the request and return sources
+with supporting evidence and retrieval limits.
+
+### Intent signature
+- Find external documentation, web information, or remote implementation examples.
+- Read a known external URL or supply external sources to another skill.
+
+### When to use
+Library documentation, web research, remote code search, known-URL reading, and
+requests whose external source channel needs selection.
+
+### When NOT to use
+- Pure local code exploration -> use configured code intelligence directly under
+  `../_shared/core/code-intelligence.md`; do not dispatch this skill.
+- Git history/blame -> `oma-scm`; architecture analysis -> `oma-architecture`.
+
+### Expected inputs
+Query or known URL; optional route hint (`docs`, `web`, `code`, `fetch`); version,
+recency, host, repository and source constraints; optional web-provider override.
+Explicit `--docs`, `--code` or `--web` selects the search route. `--gitlab` selects
+the remote code host. `--strict` requires suitable primary or official sources
+with inspected supporting content; `--wide` includes broader results with their
+limitations. These modifiers preserve the request's source constraints.
+
+### Expected outputs
+An answer or relevant source list with source URL/file reference, route, actual
+provider, source type, content-read status, supporting evidence and material
+limitations. Include publication, retrieval and version details when observed.
+Do not invent results, dates, sources or certainty.
+
+### Dependencies
+Context7 for documentation; the configured `providers.web` (`native`, `brave` or
+`exa`) for web discovery; `oma search code`/`gh`/`glab` for remote code; runtime
+retrieval tools or `oma search fetch` for known URLs. `oma search providers`
+reports registration, selection and credential metadata; `oma doctor` checks
+credential presence without verifying validity or API reachability.
+
+## Structural Flow
+
+### Routes
+| Route | Execution | Scope |
+|---|---|---|
+| `docs` | Context7 library resolution and queries; official-site discovery when needed | Official documentation within requested version constraints |
+| `web` | Selected web provider: runtime tools for `native`, CLI adapter for `brave` or `exa` | Discover URLs relevant to a query |
+| `code` | `oma search code` through the selected code host | Remote implementation matches within host/repository constraints |
+| `fetch` | Available URL reader/browser or `oma search fetch <url>` | Read a URL already supplied or discovered |
+
+### Failure and recovery
+- Missing/empty documentation lookup: use official-site discovery while retaining
+  the documentation route's version and source limits.
+- Selected web provider unavailable, unauthenticated or rate-limited: report its
+  failure. Do not silently replace an explicit provider choice with native search.
+- Search returns no candidate URLs: refine within the original constraints or
+  report no result. URL retrieval cannot replace missing search results.
+- A known URL cannot be read: use the retrieval-specific recovery reference and
+  preserve access/authentication errors and archive provenance.
+- No source meets `--strict`: report the source or evidence gap. A broader result
+  set does not automatically satisfy the strict request.
+
+### Exit
+Return supported results and disclose failed routes, unread sources, stale or
+archived content, and version limits. A provider name or domain label does not
+verify an individual claim.
+
+## Logical Operations
+
+### Canonical workflow path
+1. Resolve the requested outcome and constraints. Hand pure local-code requests
+   directly to configured code intelligence. For external work, select one primary
+   route; use `resources/intent-rules.md` only when selection is unclear.
+2. Execute that route using `resources/execution-protocol.md`. Resolve the web
+   provider from configuration or the request override. Add another route only
+   when the remaining evidence gap needs it; do not fan out by default.
+3. Read the relevant content needed to support the answer. Retain URLs and
+   inspection status, deduplicate results, and apply source/version constraints.
+   Use `resources/trust-registry.md` for provenance rules or an explicitly needed
+   optional domain diagnostic; no numeric lookup is required for ordinary results.
+4. Rank by relevance and source suitability for the question. Return the requested
+   answer or sources with evidence, the actual route/provider and material limits.
+
+### CLI entry points
+```bash
+oma search providers
+oma search web "<query>" [--provider <id>] [--limit <n>]
+oma search code "<query>" [--host gitlab] [--language <lang>] [--repo <owner/repo>]
+oma search fetch <url>
+oma search trust <domain>
+```
+Use `api`, `rss`, `meta`, `media`, `archive` or `doctor` only when the selected
+operation needs that primitive. `trust` remains an optional domain diagnostic.
+
+### Guardrails
+- Preserve explicit route, provider, source, host, version and date constraints.
+- Keep query discovery separate from reading an existing URL.
+- Treat search snippets and excerpts as limited evidence; identify unread sources.
+- Do not manufacture trust scores or promote Context7/GitHub results to claim-level
+  verification based on their provider or hosting domain.
+- Do not repeat successful routes, search local project files via the web, install
+  a provider or track a repository automatically.
+
+### Resource scope and effects
+Search may contact external services and spawn configured provider or code-host
+processes. Retrieval may launch an available browser. Query constraints and
+provenance are session context; persistent caches belong in generated state, not
+skill definitions. Local source navigation belongs to the configured code tools.
+
+## References
+- Route selection: `resources/intent-rules.md` (unclear external request)
+- Execution: `resources/execution-protocol.md` (selected route only)
+- Provenance and optional domain diagnostics: `resources/trust-registry.md` (source/evidence interpretation)
+- Recovery: `resources/error-playbook.md` (route failure)
+- Result checks: `resources/checklist.md` (applicable source and evidence checks)
+- Examples: `resources/examples.md` (unfamiliar input or boundary)
+- Local-code handoff: `../_shared/core/code-intelligence.md` (local requests)

package/skills/oma-search/resources/checklist.md

@@ -0,0 +1,36 @@
+# Search result checks
+
+## Route and scope
+
+- The route matches the requested outcome: `docs`, `web`, `code` or `fetch`.
+- Pure local-code work went directly to configured code intelligence.
+- Explicit route, provider, host, repository, version, date and source constraints
+  remain intact through recovery.
+- One primary route ran; any additional route addresses an identified evidence gap.
+- Web discovery used the configured/requested provider. An explicit provider's
+  credential or availability failure was not silently replaced by native search.
+- URL retrieval used an actual supplied/discovered URL, not a fabricated search
+  fallback target.
+
+## Evidence
+
+- Source references are real, relevant and deduplicated.
+- The actual route/provider and source type are identified when known.
+- Full-content, excerpt-only and unread-result status are distinguishable.
+- Each substantive answer claim has inspected supporting material; search snippets,
+  provider identity and hosting domains are not treated as verification.
+- Observed dates, versions, revisions and archive provenance are retained. Unknown
+  metadata remains unknown.
+
+## Modifiers and limits
+
+- `--strict` is supported by suitable primary/official content, or its evidence gap
+  is reported. No domain-score upgrade is used to make a result pass.
+- `--wide` preserves broader results with limits while retaining explicit scope.
+- Optional numeric domain diagnostics are labeled separately and copied from the
+  CLI unchanged; ordinary results do not require a score lookup.
+- Provider, access, freshness and version failures are reported. No fake results,
+  successful empty response for an auth error, or automatic retry loop is present.
+
+Return the requested answer or source list. A fixed header, numbered layout,
+score badge or decorative report template is not required.