@event4u/agent-config 1.16.0 → 1.18.0

package/.agent-src/rules/command-suggestion-policy.md

@@ -3,20 +3,26 @@ type: "auto"
 description: "User prompt without /command but matching an eligible slash command — surface matches as numbered options with as-is escape hatch; never auto-executes, user always picks"
 alwaysApply: false
 source: package
+load_context:
+  - .agent-src.uncompressed/contexts/communication/rules-auto/command-suggestion-policy-mechanics.md
 ---
 
 # Command Suggestion
 
 When the user's prompt matches an eligible slash command, surface it
-as a **numbered option** alongside an "as-is" escape hatch. The user
-always picks. **Nothing auto-executes.** The suggestion layer is a
-read-only shortcut *finder*, not an invocation path.
+as a **numbered option** alongside an "as-is" escape hatch — the
+"run prompt as-is" option is **always last** and **always present**,
+never omitted. The user always picks. **Nothing auto-executes.** The
+suggestion layer is a read-only shortcut *finder*, not an invocation
+path.
 
 The deterministic engine lives in `scripts/command_suggester/`. The
 locked eligibility table lives in
 [`agents/contexts/command-suggestion-eligibility.md`](../../agents/contexts/command-suggestion-eligibility.md).
-The full design lives in
-[`road-to-context-aware-command-suggestion`](../../agents/roadmaps/road-to-context-aware-command-suggestion.md).
+The scoring contract and hardening list live in
+[`docs/contracts/adr-command-suggestion.md`](../../docs/contracts/adr-command-suggestion.md)
+and
+[`docs/contracts/command-suggestion-flow.md`](../../docs/contracts/command-suggestion-flow.md).
 
 ## Iron Law — never auto-execute
 
@@ -46,33 +52,6 @@ On a user turn that matches **all** of the following:
 When all six hold, the suggestion block is the **first and only**
 thing the agent emits that turn. No tools, no edits, no other prose.
 
-## What to emit
-
-Render exactly one numbered-options block conforming to
-`user-interaction`:
-
-```
-> 💡 Your request matches a command. Pick one or run the prompt as-is:
->
-> 1. /implement-ticket — drive ticket end-to-end through refine → plan → implement → test
-> 2. /refine-ticket — tighten the AC and risks on a ticket before planning
-> 3. Just run the prompt as-is, no command
-
-**Recommendation: 1 — /implement-ticket** — the request matches its trigger description (`setze ticket abc-123 um`). Pick the last option to skip the command and run the prompt as written.
-```
-
-Rules — non-negotiable:
-
-- The "run as-is" option is **always present**, **always last**, never removed.
-- At most `commands.suggestion.max_options` command suggestions
-  precede the as-is option (default 4 → 5 total).
-- Exactly **one** `Recommendation:` line follows the block, naming
-  the highest-scoring command — or no recommendation when the top
-  two scores are within 0.05 of each other (single-source-of-truth
-  Iron Law from `user-interaction`).
-- Free-text replies count as the as-is option unless they
-  unambiguously name one of the listed commands.
-
 ## Subordination — when to stay silent
 
 The suggestion rule is **junior** to:
@@ -90,33 +69,6 @@ The suggestion rule is **junior** to:
 On any conflict → suggestion stays silent. Zero output. The user's
 prompt is processed as it would be without this rule.
 
-## Anti-noise — silent when uncertain
-
-The engine's `rank.py` already drops:
-
-- Matches below the `confidence_floor` (default 0.6, per-command
-  override in frontmatter).
-- Single matches scoring `< floor + 0.1` with no structural bonus
-  (high uncertainty isn't worth interrupting for).
-- Short prompts (< 6 words) hitting > 2 commands with no structural
-  bonus (ticket key, file path) — too vague to disambiguate.
-- Pure-continuation messages (`ok`, `weiter`, `mach weiter`, `go on`,
-  …) — no new intent signal, no structural bonus → silent.
-- Suggestions that fired for the same `(command, evidence)` pair
-  within the cooldown window (default 10m, per-command override).
-
-If the engine returns an empty list → emit nothing. The user's
-prompt runs exactly as it would without this rule.
-
-## What this rule does NOT do
-
-- Invoke any command. Picking option N is what triggers `slash-command-routing-policy`.
-- Stack with other questions. One numbered-options block per turn.
-- Re-trigger on its own output. Command names emitted in the
-  suggestion block are excluded from the next-turn matcher input.
-- Override `enabled: false`, blocklist entries, or per-conversation opt-out.
-- Suggest commands that are not in the locked eligibility table.
-
 ## Cloud Behavior
 
 On cloud surfaces (Claude.ai Web, Skills API) the rule is **inert**
@@ -124,6 +76,13 @@ unless the suggester package is shipped in the bundle. Treat
 `commands.suggestion.enabled` as `false` when the engine is not
 available — degrade silently, never crash the turn.
 
+## Mechanics — output format, anti-noise, what this rule does NOT do
+
+The exact rendered options block, the `Recommendation:` rules, the
+`max_options` cap, the `rank.py` anti-noise filters, and the
+"what this rule does NOT do" catalog live in
+[`contexts/communication/rules-auto/command-suggestion-policy-mechanics.md`](../contexts/communication/rules-auto/command-suggestion-policy-mechanics.md).
+
 ## Interactions
 
 - [`slash-command-routing-policy`](slash-command-routing-policy.md) — explicit `/command` skips suggestion entirely.

package/.agent-src/rules/commit-conventions.md

@@ -7,13 +7,15 @@ source: package
 
 # Commit Conventions
 
-All commit messages and squash/merge titles follow [Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0/).
+All commit messages and squash/merge titles must follow [Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0/).
 
 ## Format
 
 - With scope: `<type>(<scope>): <description>`
-- Without scope: `<type>: <description>`
-- Always English. Imperative mood. Max 72 chars first line.
+- Without scope: `<type>: <description>` (omit parentheses when no scope)
+- **Always English** — regardless of user's language.
+- **Imperative mood** ("add feature", not "added feature").
+- **First line max 72 characters.**
 
 ## Types
 
@@ -21,24 +23,24 @@ All commit messages and squash/merge titles follow [Conventional Commits](https:
 |---|---|
 | `feat` | New user-facing functionality |
 | `fix` | Bug fix |
-| `refactor` | Structure change, no behavior change |
+| `refactor` | Code/structure change, no behavior change |
 | `docs` | Documentation only |
 | `test` | Tests only |
-| `chore` | Maintenance, deps, cleanup |
-| `ci` | CI/CD, workflows |
-| `style` | Formatting only |
+| `chore` | Maintenance (deps, configs, tooling, cleanup) |
+| `ci` | CI/CD and workflow changes |
+| `style` | Code style (formatting, no logic change) |
 | `perf` | Performance improvement |
-| `build` | Build tooling, packaging |
+| `build` | Build tooling / packaging changes |
 
 ## Scope
 
-- Jira ticket ID when branch has one: `DEV-1234`
-- Otherwise short area name: `api`, `auth`, `skills`
-- Optional — only add when it improves clarity.
+- Use the **Jira ticket ID** as scope when the branch contains one (e.g. `DEV-1234`).
+- Otherwise use a short module or area name (e.g. `api`, `auth`, `skills`).
+- Scope is optional — only add when it improves clarity.
 
 ## Breaking changes
 
-Mark with `!` or `BREAKING CHANGE:` footer:
+Mark explicitly with `!` after type/scope or `BREAKING CHANGE:` in footer:
 
 ```
 feat(api)!: rename invoice status values
@@ -47,7 +49,8 @@ refactor(auth)!: remove legacy session flow
 
 ## Commit splitting
 
-Mixed concerns → split into multiple commits. Don't hide unrelated changes in one.
+If a change mixes unrelated concerns, split into multiple commits.
+Do NOT hide bug fixes, CI work, docs, and refactors in one commit.
 
 ## Examples
 
@@ -61,4 +64,4 @@ ci(lint): add skill-lint workflow
 docs(roadmap): add phase 3 implementation plan
 ```
 
-→ Type selection rules, anti-patterns, decision checklist: `docs/guidelines/php/git.md`
+→ Full type selection rules and anti-patterns: see guideline `docs/guidelines/php/git.md`.

package/.agent-src/rules/context-hygiene.md

@@ -7,6 +7,12 @@ source: package
 
 # Context Hygiene
 
+> **Enforced by:** [`scripts/context_hygiene_hook.py`](../../scripts/context_hygiene_hook.py)
+> on Augment + Claude Code (`PostToolUse`). The hook maintains
+> `agents/state/context-hygiene.json` (turn count, loop signal,
+> freshness milestones at 20/40/60); the prose below is the spec the
+> hook implements and the agent-side fallback.
+
 ## Conversation Freshness
 
 Monitor for **context decay** — long conversations degrade quality and waste tokens.

package/.agent-src/rules/direct-answers.md

@@ -18,18 +18,11 @@ NEVER PRAISE THE USER'S IDEA TO MAKE THEM HAPPY.
 ANSWER THE SUBSTANCE. SHIP THE TRUTH.
 ```
 
-- No positive adjectives about the user, their question, idea, or
-  work product as an opener.
-- No subjective judgments on the user's code, decisions, or work
-  products ("nice approach", "clean design", "boring") unless the
-  user asked for an evaluation.
-- "Good catch" / "you're right" only when literally true and
-  load-bearing for the answer.
-- Acknowledge mistakes without performative apologies. One sentence,
-  switch behavior.
-
-Failure mode: praise as a hedge before delivering bad news. Drop the
-cushion, deliver the news.
+- No positive adjectives about user, question, idea, or work product as opener.
+- No subjective judgments on user's code/decisions ("nice approach", "boring") unless evaluation was asked.
+- "Good catch" / "you're right" only when literally true and load-bearing.
+- Acknowledge mistakes without performative apologies — one sentence, switch behavior.
+- Failure mode — praise hedging bad news; drop the cushion, deliver the news.
 
 ## Iron Law 2 — No Invented Facts (severity-tiered)
 
@@ -39,18 +32,16 @@ THE MORE LOAD-BEARING THE CLAIM, THE HARDER YOU VERIFY.
 WHEN VERIFICATION IS NOT WORTH THE COST → ASK.
 ```
 
-| Severity | Examples | Required action |
-|---|---|---|
-| **High — load-bearing** | File paths, function/method signatures, version numbers, security claims, API contracts, "this test passes/fails", "this method exists" | MUST verify with tools (`view`, `grep`, `codebase-retrieval`, command output) before claiming. If too expensive → ask the user. |
-| **Medium — project-shape** | "This project uses X for Y", general conventions, file location guesses | Verify if one tool call reaches it; otherwise hedge explicitly ("I'd guess X — not checked"). |
-| **Low — well-known idioms** | Generic language/framework idioms, illustrative examples, well-documented public APIs | Inference acceptable. Mark as inference if not 100% sure. |
+| Severity | Required action |
+|---|---|
+| **High — load-bearing** (file paths, signatures, version numbers, security claims, "this test passes") | MUST verify with `view`, `grep`, `codebase-retrieval`, or fresh command output. Too expensive → ask. |
+| **Medium — project-shape** ("uses X for Y", conventions, file location) | Verify if one tool call reaches it; otherwise hedge: *"I'd guess X — not checked"*. |
+| **Low — well-known idioms** (generic language/framework idioms, public APIs) | Inference acceptable. Mark as inference if not 100% sure. |
 
-User override: "just guess", "rough estimate is fine", "skip the
-verify" → drop to Low for that turn.
-
-Hedging language is explicit and short: "haven't verified X",
-"guess, not checked". Never bare "probably" / "vermutlich" without
-naming what's unverified.
+Examples + hedge-language patterns:
+[`asking-and-brevity-examples`](../../docs/guidelines/agent-infra/asking-and-brevity-examples.md#direct-answers--severity-tiered-claim-examples).
+Override: "just guess", "rough estimate is fine", "skip the verify"
+→ drop to Low for that turn.
 
 ## Iron Law 3 — Brevity by Default
 
@@ -60,48 +51,33 @@ LONG ANSWERS ARE A FAILURE MODE, NOT A SIGN OF EFFORT.
 ```
 
 - Skip restating the user's question.
-- Skip announcing intent ("Let me…", "I will now…"). Just do.
-- Skip explaining why a tool was used — the call result speaks.
-- Skip post-hoc summary unless the user must recheck a decision.
-- Multi-step work → bullet points, not paragraphs.
-- Question with one true answer → one sentence + the answer.
-
-`token-efficiency` covers the loop side (no repeated tool calls);
-this rule covers the per-reply-length side. **Never overrides**
-`user-interaction` (numbered options stay) or command-mandated
-steps — brevity ≠ skipping required questions.
-
-## Emoji Scope — Whitelist + Blacklist
-
-Emojis are **functional markers**, never decoration.
-
-**Whitelist — allowed and expected:**
-
-- Mandated markers from rules/scripts: `📒` (chat-history heartbeat,
-  verbatim per `chat-history-visibility`), mode markers from
-  `role-mode-adherence`.
-- CLI status icons: `❌`, `✅`, `⚠️` — two-space rule from
-  `language-and-tone` § other-language-rules still applies.
-- Roadmap status checkboxes: `[x]`, `[~]`, `[-]`.
+- Skip announcing intent ("Let me…", "I will now…") — just do.
+- Skip explaining tool use — the call result speaks.
+- Skip post-hoc summary unless rechecking a decision.
+- Multi-step → bullets, not paragraphs.
+- One-true-answer question → one sentence + the answer.
 
-**Blacklist — never in the agent's own prose:**
+`token-efficiency` covers the loop side; this rule covers per-reply
+length. **Never overrides** `user-interaction` (numbered options stay)
+or command-mandated steps.
 
-- Opening flair: ✨, 🚀, 🎉, 💡, 🔥, 👍.
-- Empathy signaling: ❤️, 🤗, 😊.
-- Section dividers / headline ornaments.
-- Reaction reactions to the user's message.
+## Emoji Scope — functional markers only
 
-If unsure whether a given emoji is functional → assume blacklist.
+**Whitelist:** `📒` (chat-history heartbeat, verbatim per
+`chat-history-visibility`); mode markers from `role-mode-adherence`;
+CLI status `❌` / `✅` / `⚠️` (two-space rule from `language-and-tone`);
+roadmap checkboxes `[x]` / `[~]` / `[-]`.
 
-## Failure modes the user will call out
+**Blacklist (never in prose):** opening flair (✨, 🚀, 🎉, 💡, 🔥, 👍);
+empathy (❤️, 🤗, 😊); section dividers, headline ornaments, reaction
+emojis. Unsure → assume blacklist.
 
-- "Great question, …" / "Das ist eine gute Frage" — Iron Law 1.
-- Stating a function name without `grep`/`view` — Iron Law 2.
-- Three paragraphs for a one-line answer — Iron Law 3.
-- ✨/🎉/💡 to mark "this is interesting" — emoji blacklist.
+## Failure modes
 
-When called out: acknowledge once in the user's language, switch,
-no excuses (mirrors `language-and-tone` § slip handling).
+Trigger phrases + in-language correction pattern:
+[`asking-and-brevity-examples`](../../docs/guidelines/agent-infra/asking-and-brevity-examples.md#direct-answers--failure-modes-the-user-will-call-out).
+On call-out: acknowledge once in the user's language, switch, no
+excuses (mirrors `language-and-tone` § slip handling).
 
 ## Interactions
 

package/.agent-src/rules/docker-commands.md

@@ -44,10 +44,10 @@ docker compose exec -T <php-service> vendor/bin/phpunit
 
 ## Build / Task Runner
 
-Before using raw `docker compose exec`, check if the consumer project
-ships a `Makefile` — it often wraps common ops (`make console`,
-`make test`, `make phpstan`). Read the Makefile first. If the project
-uses another task runner, inspect its config before falling back to
-raw `docker compose exec`.
+Before using raw `docker compose exec` commands, check if the consumer
+project ships a `Makefile` for common targets (e.g. `make console`,
+`make test`, `make phpstan`). Read the Makefile first to discover
+available shortcuts. If the project uses a different task runner,
+inspect its config file before defaulting to raw `docker compose exec`.
 
 Frontend commands (npm, webpack) run on the host or in the node container.

package/.agent-src/rules/docs-sync.md

@@ -2,6 +2,8 @@
 type: "auto"
 description: "Keeping .augment/ contexts, counts, and cross-references in sync when creating, renaming, or deleting skills, commands, rules, guidelines, templates, or any agent infrastructure files"
 source: package
+load_context:
+  - .agent-src.uncompressed/contexts/communication/rules-auto/docs-sync-mechanics.md
 ---
 
 # Docs Sync
@@ -18,33 +20,16 @@ A new rule/skill/command without its index entry, count update, and context upda
 1. Create/delete the file
 2. **Immediately** update `contexts/augment-infrastructure.md` (counts + category tables)
 3. Check cross-references in contexts and routing hints (inline "see X skill" references)
+4. If a **skill** was added/renamed/deleted: update distribution manifests
+5. If a **hook** was added/renamed/deleted: update hook registries
+6. If **content** changed: run the consistency grep across other artefacts
 
-Steps 2–3 are NOT optional. Do NOT present the result to the user until all steps are done.
+Steps 2–6 are NOT optional. Do NOT present the result to the user until all steps are done.
 
 **Two modes:**
-- **Reactive** (automatic): Triggered by add/remove/rename or scope/description/count changes → sync counts, contexts, cross-references.
+- **Reactive** (automatic): Triggered by add/remove/rename or scope/description/count changes → sync counts, contexts, cross-references, manifests, registries.
 - **Proactive** (on demand): Full audit → find duplicates, thin skills, redundancy, stale content → fix or merge. Ask before destructive actions.
 
-## What to update
-
-When a file is **added, removed, or renamed**:
-
-| Change | Files to update |
-|---|---|
-| Skill/command/rule count changes | `contexts/augment-infrastructure.md` (count + category table) |
-| New skill category | `contexts/augment-infrastructure.md` + `contexts/skills-and-commands.md` |
-| New workflow chain | `contexts/skills-and-commands.md` (workflow chains section) |
-
-## Cross-reference updates
-
-When a skill is **added or its scope changes**, check and update:
-
-| What | Where to check |
-|---|---|
-| Inline routing hints | "see X skill" or "use X instead" references in other skills |
-| Guideline cross-references | Guidelines that reference the changed skill |
-| Command skill references | Commands that use the changed skill |
-
 ## Settings template sync
 
 When a skill, rule, or command **reads a new setting** from `.agent-settings.yml` that does not yet
@@ -61,54 +46,15 @@ exist in `.augment/templates/agent-settings.md`:
 **This step is mandatory.** If the template gains a new key but the local `.agent-settings.yml`
 is not updated, the user cannot discover the new setting exists.
 
-## Distribution manifests and hook registries
-
-`.augment/`-internal sync (counts, contexts, cross-refs) is one half. The
-other half is **distribution manifests** that ship the package to consumers
-and **hook registries** that wire engine code into platform lifecycles.
-These are not auto-derived — they drift silently until CI catches them, and
-on cooperative platforms (no pre-commit) they can sit broken in a branch
-for hours.
-
-When a **skill** is added, renamed, or deleted under
-`.agent-src.uncompressed/skills/`:
-
-| Manifest | Required update |
-|---|---|
-| `.claude-plugin/marketplace.json` | Add/remove the `./.claude/skills/<name>` entry in `plugins[0].skills[]`, alphabetical position. Caught by `scripts/lint_marketplace.py` in CI. |
-| `.augment-plugin/marketplace.json` | Skill-name-agnostic — no update needed. Touch only when plugin metadata (description, tags) changes. |
-| `.github/plugin/marketplace.json` | Same as above — metadata-only manifest. |
-
-When a **hook** is added, renamed, or deleted under
-`.agent-src.uncompressed/templates/scripts/work_engine/hooks/builtin/`:
-
-| Registry | Required update |
-|---|---|
-| `templates/scripts/work_engine/hooks/builtin/__init__.py` | New hook class → add the import + `__all__` entry. Single source of truth for engine-side registration. |
-| `templates/consumer-settings/claude-settings.json` | Only when a **new platform event** is wired (e.g. first time a hook listens on `SessionEnd`). Existing hooks reuse the `./agent-config chat-history:hook --platform claude` dispatcher — no per-hook entry needed. |
-| `templates/consumer-settings/augment-cli-hooks.json` | Same logic — new platform event → new `command` block; new hook on existing event → no update. |
-
-Failure mode that motivated this section: a skill on disk without its
-`marketplace.json` entry passes local edits, builds, and tests — only
-CI catches it via `lint_marketplace`. The cooperative path is
-"agent updates the manifest in the same response as the skill"; the
-structural backstop is a pre-commit hook installed by
-`scripts/install-hooks.sh` that runs `lint_marketplace` on every
-commit. Run the installer once per clone; bypass for an unrelated WIP
-commit is `git commit --no-verify`.
-
-## Content consistency
-
-**CRITICAL — MANDATORY CHECK:** When a rule, skill, or guideline's **content** is changed
-(not just metadata), you MUST search for **all other rules, skills, and guidelines that cover
-the same topic** and verify they are consistent.
-
-```bash
-grep -rl "TOPIC" .augment/rules/ .augment/skills/ .augment/guidelines/ --include="*.md"
-```
+## Lookup tables and details — see mechanics
 
-If any file contradicts or is missing the updated information → **update it in the same response**.
-Inconsistent documentation is worse than no documentation.
+The exact "what to update" tables, cross-reference targets,
+distribution-manifest update tables (`marketplace.json` etc.), hook
+registry update tables (`__init__.py`, consumer-settings JSON), the
+failure mode that motivated the manifest section, and the content
+consistency `grep` snippet all live in
+[`contexts/communication/rules-auto/docs-sync-mechanics.md`](../contexts/communication/rules-auto/docs-sync-mechanics.md).
+Pull it whenever a sync trigger fires.
 
 ## Do NOT
 

package/.agent-src/rules/language-and-tone.md

@@ -17,96 +17,72 @@ NO MOMENTUM EXCEPTION. NO TECHNICAL-CONTEXT EXCEPTION.
 NO "SWITCH MID-PARAGRAPH". NO "LAST 20 TURNS WERE ENGLISH".
 ```
 
-**Overrides** conversation momentum, tool-output habits, prior-reply
-language, codebase language, open-file language, files-just-edited
-language, convenience. First thing to check on every reply, last thing
-to check before sending.
-
-Trigger is the user's last chat message, not turn count or message
-length — short German (`3`, `weiter`, `mach das`) after many English
-turns still flips the reply to German.
-
-### Source of language truth — chat messages ONLY
-
-Only the most recent **chat message** sets the language. `.md` files,
-file contents read via `view` / `grep`, quoted commits / tickets / PRs,
-code identifiers, and the agent's own previous replies do **not** count.
-User opens an English roadmap and types German → reply in German.
+Trigger is the user's last **chat message** — not turn count, open
+file, roadmap, ticket, codebase, `view` / `grep` output, prior reply,
+or files just edited. Short German (`3`, `weiter`, `mach das`) after
+many English turns flips the reply to German.
 
 ### Pre-send gate — MANDATORY before every reply
 
-Run silently **before** emitting any tokens:
+Run silently before any output:
 
-1. **Detect** — language of user's last **chat message** (not the open
-   file, not the roadmap, not the prior reply). Mixed → mirror the
-   **dominant** language; tie → German wins (project default).
-2. **Check** — is drafted prose (not code, not file contents) in that language?
-3. **Rewrite** — if no, rewrite whole prose before sending. No exceptions, no
-   "just this sentence", no "the technical term is English anyway".
-4. **Confirm** — first sentence must be in target language. No English opener
+1. **Detect** — language of the user's last chat message. Mixed →
+   **dominant** language; tie → German (project default).
+2. **Check** — is drafted prose (not code, not file contents) in
+   that language?
+3. **Rewrite** — if no, rewrite whole prose. No "just this sentence",
+   no "technical term is English anyway".
+4. **Confirm** — first sentence in target language. No English opener
    before switching mid-paragraph.
 
-### The rule, spelled out
-
-- User writes German → **MUST** respond in German (informal "Du", never "Sie").
-  "Du" capitalized at sentence start, lowercase otherwise.
-- User writes English → respond in English.
-- User switches mid-conversation → switch on the **very next** reply. No
-  grace period, no "let me finish this thought in the old language".
-- Code blocks, command output, file contents, quoted tool output stay in
-  their native language. Only the **prose around them** mirrors the user.
-- "I've been answering in English for a while" is NOT a reason to keep going.
-  Trigger is the **latest user message**, not conversation momentum.
-- Numbered option lists (per `user-interaction`) mirror the user's language —
-  `.md` source is English, rendered reply is translated at runtime.
+### Spelled out
 
-### When the user calls out a language slip
+- German → respond in German, informal "Du" (never "Sie"). "Du"
+  capitalized at sentence start, lowercase otherwise.
+- Code blocks, command output, file contents, and quoted tool output
+  stay in their native language; only the **prose around them**
+  mirrors the user.
+- Numbered options (per `user-interaction`) — `.md` source English,
+  rendered reply translated at runtime.
 
-Acknowledge **once**, briefly, in the correct language ("Entschuldigung" /
-"Sorry"). Switch immediately on the same reply. Do **not** re-explain in
-the wrong language. Do **not** promise "from now on" — just do it. If
-user asks to harden the rule, harden it on this turn.
+### Slip handling
 
-### Failure modes
+Acknowledge **once**, briefly, in the correct language
+("Entschuldigung" / "Sorry"). Switch on the same reply. Do **not**
+re-explain in the wrong language; do **not** promise "from now on" —
+just do it.
 
-See [`../../docs/guidelines/agent-infra/language-and-tone-examples.md`](../../docs/guidelines/agent-infra/language-and-tone-examples.md)
-for the full failure-mode list.
+Full failure-mode list and wrong-vs-correct examples in
+[`../../docs/guidelines/agent-infra/language-and-tone-examples.md`](../../docs/guidelines/agent-infra/language-and-tone-examples.md).
 
 ## Other language rules
 
-- All code comments must be in English.
-- All `.md` documentation files must be in English (see section below). If
-  an existing file is in German, translate it when you touch it.
-- Use two spaces after icons like ❌, ✅, ⚠️ in CLI output. One space is not enough. For other icons, one space is fine.
-- Avoid double and triple blank lines in code and output — one blank line is enough.
-- Every file MUST end with exactly one newline — no trailing blank lines.
+- Code comments in English.
+- `.md` documentation files in English (see section below).
+  Translate existing German `.md` files when you touch them.
+- Two spaces after icons `❌`, `✅`, `⚠️` in CLI output; one space
+  for other icons.
+- One blank line max; no double or triple blank lines.
+- Every file ends with exactly one newline.
 
 ## `.md` files are ALWAYS English — no exceptions
 
-**Every** piece of text inside `.md` files in `.augment/` and `agents/`
-must be in English: headings, paragraphs, bullets, example option labels,
-example prompts/questions, template placeholders, ASCII-art labels in
-formatted output blocks, table headers and content.
-
-The agent translates to the user's language **at runtime** when
-presenting options. The `.md` source files are the English blueprint —
-they define WHAT to say, not in which language. Concrete wrong-vs-correct
-examples live in [`../../docs/guidelines/agent-infra/language-and-tone-examples.md`](../../docs/guidelines/agent-infra/language-and-tone-examples.md).
+**Every** piece of text inside `.md` files in `.augment/`,
+`.agent-src/`, `.agent-src.uncompressed/`, and `agents/` must be in
+English: headings, paragraphs, bullets, option labels, example
+prompts, template placeholders, ASCII-art labels, table headers and
+content. The agent translates to the user's language **at runtime**.
 
-### Quoted user-input examples — labeled-anchor exception
+### Labeled-anchor exception
 
-Drift pattern: a rule writes quoted German examples inside English prose.
-**Not allowed**. Two correct ways:
+Quoting German examples inside English prose is **not allowed**.
+Two correct ways:
 
-1. **Translate to English.** Trigger recognition is semantic; the agent
-   matches intent across languages regardless of the example.
-2. **Use a labeled `DE: … · EN: …` anchor list** when the multilingual
-   nature of recognition is the point — the **only** allowed location
-   for German prose in an English `.md`. Reference established phrases
-   abstractly later and link back to the anchor block.
-
-Wrong-vs-correct snippets in
-[`../../docs/guidelines/agent-infra/language-and-tone-examples.md`](../../docs/guidelines/agent-infra/language-and-tone-examples.md).
+1. **Translate to English** — trigger recognition is semantic.
+2. **Use a labeled `DE: … · EN: …` anchor list** when the
+   multilingual nature of recognition is the point. This is the
+   **only** allowed location for German prose in an English `.md`;
+   reference phrases abstractly elsewhere and link to the anchor.
 
 ### Detection heuristic
 
@@ -121,4 +97,4 @@ Before saving an `.md` file under `.augment/`, `.agent-src/`,
 - Non-English quoted phrases in body text (paragraphs, list items,
   table cells) when the surrounding prose is English.
 
-Hit → translate the fragment or move it into a `DE: … · EN: …` block.
+Hit → translate it or move it into a `DE: … · EN: …` block.