@event4u/agent-config 9.16.0 → 9.17.0

package/.claude-plugin/marketplace.json

@@ -6,7 +6,7 @@
   },
   "metadata": {
     "description": "Bootstrap shim for the event4u agent-config suite. This plugin ships ONLY the dispatcher hooks plus one install-pointer skill — ALL content (skills, rules, commands) is installed via `npx -y @event4u/agent-config init`, which writes the file projection AND registers the same hooks in ~/.claude/settings.json. DEPRECATED as a content channel; kept so the marketplace listing stays discoverable and existing installs keep working hooks.",
-    "version": "9.16.0",
+    "version": "9.17.0",
     "keywords": [
       "agent-config",
       "skills",

package/AGENTS.md

@@ -24,7 +24,7 @@ task ci                # full pipeline — green before PR
 - **Content pipelines** (A→D): [`docs/architecture.md`](docs/architecture.md).
 - **Editing this repo**: [`source-of-truth`](src/rules/source-of-truth.md) + [`agents-md-thin-root`](src/skills/agents-md-thin-root/SKILL.md).
 - **Consumer story** — `npx` + `scripts/install.sh`: [`README.md`](README.md).
-- **Personas** — 11 review lenses (6 + 5): [`docs/personas.md`](docs/personas.md).
+- **Personas** — 34 lenses (5 core + 24 specialists + 5 advisors): [`docs/personas.md`](docs/personas.md).
 - **Discovery** — workspaces/packs: [`ADR-013`](docs/decisions/ADR-013-discovery-frontmatter-contract.md) + [`customization`](docs/customization.md#workspaces--packs-discovery).
 - **Root & `agents/` layout** — `src/` = source ([`ADR-050`](docs/decisions/ADR-050-workspace-vs-package-root-boundary.md)); `agents/` contract: [`agents-layout`](docs/contracts/agents-layout.md).
 - **Security** — [`SECURITY.md`](SECURITY.md); [`docs/threat-model.md`](docs/threat-model.md).

package/CHANGELOG.md

@@ -262,6 +262,59 @@ Entry-shape contract: [`docs/contracts/CHANGELOG-conventions.md`](docs/contracts
 > that forces a new era split (`# Era: 9.16.x`, etc.) — see
 > [`docs/contracts/CHANGELOG-conventions.md § Era splits`](docs/contracts/CHANGELOG-conventions.md).
 
+## [9.17.0](https://github.com/event4u-app/agent-config/compare/9.16.0...9.17.0) (2026-08-04)
+
+### Release highlights
+
+<!-- Curated head: fill before merge, keep it under 10 lines, and leave `_none_` where it is genuinely the answer. The generated log below is unchanged. -->
+- **Behaviour changes:** `/optimize:deep` is now bounded — plan-only with a schema-backed limits block, a 5-loop ceiling, a pre-registered target metric, and hard exclusions for kernel rules and stable contracts (980b4f1).
+- **Default changes + migration:** `/optimize:deep` defaults to plan-only; push/PR/loop execution now requires an explicit `--mode=execute`. No data or config migration.
+- **Security and correctness:** symlink traversal is confined in the catalog/count walkers (514189a); `ip-address` and `fast-uri` bumped past high npm-audit advisories (be9fb0f, 22887ba).
+- **Honest nulls:** no new nulls in this span — the highlight-plausibility gate itself ships here, and the false-`_none_` 9.13.0/9.14.0 heads were backfilled retroactively (491f98b).
+- **Known limitations:** _none_
+
+### Features
+
+* **agent-handoff:** CLI-backed resume flow as the primary command surface ([0ca1ee9](https://github.com/event4u-app/agent-config/commit/0ca1ee9136263e03eff3a62aed83b4cc70aed36f))
+* **cli:** agent-config handoff — picker, launch adapters, print mode ([d88eea9](https://github.com/event4u-app/agent-config/commit/d88eea976c2752f67c6d342011904a664d9e4329))
+* **handoff:** session enumeration + deterministic handoff generator ([55d2459](https://github.com/event4u-app/agent-config/commit/55d24597150c1155379d22e37593d1c78fca3826))
+* **hooks:** one-shot handoff-context session_start concern ([9fa684d](https://github.com/event4u-app/agent-config/commit/9fa684d62800b21871b09582fb2c6f109ea03166))
+* **commands:** bound /optimize:deep — plan-only default, enforced limits ([980b4f1](https://github.com/event4u-app/agent-config/commit/980b4f18319db38b0a2c4d9bd2ab0783c17de672))
+* **release:** finding-disposition ledger with stable ids ([4428e57](https://github.com/event4u-app/agent-config/commit/4428e5788f5cc8ef6c98da1e05fd6879374f3fb8))
+* **release:** highlight plausibility gate + backfill the 9.13.0/9.14.0 heads ([491f98b](https://github.com/event4u-app/agent-config/commit/491f98b32767f5c84e0d46f5d2938fa298fb0f4b))
+* **release:** derive all four release surfaces from the changelog at head ([d947247](https://github.com/event4u-app/agent-config/commit/d9472472de75755583fd6220975e602804f46096))
+* **release:** drill — run the real step machinery against a simulated git/gh world ([a937e59](https://github.com/event4u-app/agent-config/commit/a937e5908edcf3953fc73a6a552de1a51fdc98cd))
+
+### Bug Fixes
+
+* **agent-handoff:** trim suggestion.trigger_context under the 240-char schema cap ([708f963](https://github.com/event4u-app/agent-config/commit/708f963e3bf674f9ec1449673608fbf413897641))
+* **deps:** bump ip-address past the SSRF advisory trio (npm audit high) ([be9fb0f](https://github.com/event4u-app/agent-config/commit/be9fb0f1f58c39570deb38ab317d159c657fc14b))
+* **deps:** bump fast-uri past GHSA-7p8r-x3mc-p8w7 (npm audit high) ([22887ba](https://github.com/event4u-app/agent-config/commit/22887badc0e18f7e3fb7301e9aa02465e02e90e5))
+* **ci:** gate the findings-artifact upload on the file, not hashFiles ([7779d55](https://github.com/event4u-app/agent-config/commit/7779d55efb79b98f7ee6dabd84ed1659db6b4ec6))
+* **security:** confine symlink traversal in the catalog/count walkers ([514189a](https://github.com/event4u-app/agent-config/commit/514189a18f44d55cd45cb6a2b5a7236365e4d0a3))
+
+### Documentation
+
+* **roadmap:** close and archive road-to-package-renewal (central) ([01b7ff6](https://github.com/event4u-app/agent-config/commit/01b7ff66b7886dfed9362682fc1262b15145adaa))
+* **handoff:** regenerate index catalogs ([a12e89e](https://github.com/event4u-app/agent-config/commit/a12e89e4f56c2bac0877f6f87c79a7ebb7168210))
+* **handoff:** sync generated catalogs, contracts, and archive the roadmap ([3505bb5](https://github.com/event4u-app/agent-config/commit/3505bb522fc69d2fdc9c5c3f4abc4ab35c3727ca))
+* **roadmap:** close and archive road-to-release-truth ([e8c1200](https://github.com/event4u-app/agent-config/commit/e8c120005250cfa116194b2aad54e237a8e09d8e))
+
+### Tests
+
+* **release:** script the release-truth surface interactions in the drill ([6ce8025](https://github.com/event4u-app/agent-config/commit/6ce8025597c0502f25b4cd350f69d20424568d1c))
+
+### CI
+
+* **release:** wire surface-equality, highlight-plausibility, finding-dispositions ([86866ed](https://github.com/event4u-app/agent-config/commit/86866edc918f4fbe3f2b2a1f6047584e2bc78b44))
+
+### Chores
+
+* ignore worktrees ([1c97785](https://github.com/event4u-app/agent-config/commit/1c9778549a4ea2dcb5c9a63b1e1252e5584876ce))
+* **budgets:** cli_help_command_count 81 -> 82 for the handoff subcommand ([b27e14d](https://github.com/event4u-app/agent-config/commit/b27e14d2a71ec4e682fbae34eec4b710d0165022))
+
+Tests: 10692 (+104 since 9.16.0)
+
 ## [9.16.0](https://github.com/event4u-app/agent-config/compare/9.15.0...9.16.0) (2026-08-03)
 
 ### Release highlights

package/README.md

@@ -4,7 +4,7 @@
 
 [![Smoke](https://github.com/event4u-app/agent-config/actions/workflows/smoke.yml/badge.svg)](https://github.com/event4u-app/agent-config/actions/workflows/smoke.yml) [![Public install smoke (3 OS × 2 Node)](https://github.com/event4u-app/agent-config/actions/workflows/smoke-public-install.yml/badge.svg)](https://github.com/event4u-app/agent-config/actions/workflows/smoke-public-install.yml) [![npm](https://img.shields.io/npm/v/@event4u/agent-config?style=flat-square&label=npm&color=orange)](https://www.npmjs.com/package/@event4u/agent-config) [![agent-config MCP server](https://glama.ai/mcp/servers/event4u-app/agent-config/badges/score.svg)](https://glama.ai/mcp/servers/event4u-app/agent-config) [![MCP Toplist](https://mcptoplist.com/badge/glama%2Fevent4u-app%2Fagent-config.svg)](https://mcptoplist.com/server/glama%2Fevent4u-app%2Fagent-config)
 
-[![Skills](https://img.shields.io/badge/Skills-288-orange?style=flat-square)](dist/agent-src/skills/) [![Rules](https://img.shields.io/badge/Rules-111-orange?style=flat-square)](dist/agent-src/rules/) [![Commands](https://img.shields.io/badge/Commands-193-orange?style=flat-square)](dist/agent-src/commands/) [![Guidelines](https://img.shields.io/badge/Guidelines-101-orange?style=flat-square)](docs/guidelines/) [![Personas](https://img.shields.io/badge/Personas-29-orange?style=flat-square)](dist/agent-src/personas/) [![Advisors](https://img.shields.io/badge/Advisors-5-orange?style=flat-square)](dist/agent-src/personas/advisors/)
+[![Skills](https://img.shields.io/badge/Skills-288-orange?style=flat-square)](dist/agent-src/skills/) [![Rules](https://img.shields.io/badge/Rules-111-orange?style=flat-square)](dist/agent-src/rules/) [![Commands](https://img.shields.io/badge/Commands-193-orange?style=flat-square)](dist/agent-src/commands/) [![Guidelines](https://img.shields.io/badge/Guidelines-102-orange?style=flat-square)](docs/guidelines/) [![Personas](https://img.shields.io/badge/Personas-29-orange?style=flat-square)](dist/agent-src/personas/) [![Advisors](https://img.shields.io/badge/Advisors-5-orange?style=flat-square)](dist/agent-src/personas/advisors/)
 
 **Try it in 30 seconds** — drop one read-only subagent into any repo and watch it gate "done": `@production-validator check this branch is actually done`. No wizard, no lock-in, nothing else installed — the [30-second wedge ↓](#quickstart) is the whole first step. Start at the proof, not the catalog: **[event4u-app.github.io/agent-config/proof/](https://event4u-app.github.io/agent-config/proof/)**.
 

package/dist/agent-src/commands/agent-handoff.md

@@ -2,18 +2,18 @@
 model_tier: medium
 name: agent-handoff
 pack: meta
-intent: "Generate a session-handoff summary to continue work in a fresh chat"
+intent: "Resume a previous session in a fresh chat: pick a session, generate a handoff, auto-seed the next session"
 routes_to: [agent-docs-writing]
 replaces: []
 tier: 0
 visibility: visible
 skills: [agent-docs-writing]
-description: Generate a context summary for continuing work in a fresh chat. Replaces the session system.
-argument-hint: "[with tasks]"
+description: Pick a recent session, generate a handoff from its transcript, and seed a fresh session with it — or summarize the live conversation for copy-paste.
+argument-hint: "[--print | --file | with tasks]"
 suggestion:
   eligible: true
-  trigger_description: "user asks for an agent handoff, fresh-chat summary, or context-summary to paste into a new chat"
-  trigger_context: "explicit verbatim ask — 'agent handoff', 'agend handoff' (typo), 'fasse für neuen chat zusammen', 'handoff summary', 'context summary for fresh chat'. Never inferred without a direct handoff / summary intent."
+  trigger_description: "user asks for an agent handoff, to resume/continue a previous session in a fresh chat, or a context-summary to paste into a new chat"
+  trigger_context: "explicit ask — 'agent handoff'/'agend handoff' (typo), 'handoff summary', 'context summary for fresh chat', 'resume/continue an old session in a fresh chat', 'fasse für neuen chat zusammen', 'neue session mit altem stand'. Never inferred."
 workspaces:
   - agent-config-maintainer
 packs:
@@ -22,24 +22,63 @@ packs:
 
 # /agent-handoff
 
-Generates a concise handoff prompt the user can paste into a new chat.
-This replaces the old session system — no files to save/load, just a copy-paste prompt.
+Resume-style flow: pick one of the recent sessions, a handoff is generated
+deterministically from its transcript, and the next session starts with that
+handoff already injected as context. The legacy copy-paste summary of the
+*live* conversation stays available as `--print`.
 
-## Steps
+## Steps — primary flow (CLI-backed session picker)
 
-### 1. Gather context
+### 1. List the recent sessions
 
-Collect the following from the current conversation:
+```bash
+agent-config handoff --list
+```
+
+Sources, merged newest-first: the cross-host chat-history log (primary),
+native Claude Code transcripts (`~/.claude/projects/<slug>/`), and the Codex
+session store (`~/.codex/sessions/`, Codex's only source).
 
-- **Branch**: `git branch --show-current`
-- **Uncommitted changes**: `git status --short`
-- **Recent commits**: `git log --oneline -5`
-- **Active roadmap**: Check `agents/roadmaps/` for any active roadmap
-- **Task list**: Only include if the user explicitly asks for it (e.g., "/agent-handoff with tasks")
+### 2. Present the picker
 
-### 2. Generate handoff prompt
+Show the sessions as ONE numbered-options block (per `user-interaction` —
+date · branch · summary per line) and let the user pick. One question, then
+wait.
 
-Create a fenced code block the user can copy-paste into a new chat:
+### 3. Generate the handoff for the pick
+
+```bash
+agent-config handoff --session <id>
+```
+
+The generator extracts the handoff sections deterministically from the
+transcript — no LLM spend, reproducible, privacy-floored per line (violating
+lines are DROPPED and counted, never rewritten) — and writes the result
+atomically to `agents/runtime/state/handoff-context.md` (gitignored runtime
+state, one-shot).
+
+### 4. Hand over to the fresh session
+
+Tell the user: **start a new session — the handoff is injected
+automatically** (the `handoff-context` session_start hook consumes the file
+exactly once, then deletes it). In a terminal-capable context, offer
+`agent-config handoff --session <id> --launch claude` (or `--launch codex`)
+to spawn the fresh session directly:
+
+- `claude` — seeds via the session_start hook (clean first message).
+- `codex` — no hook surface there; the adapter passes the handoff as the
+  initial prompt instead. Same UX, different transport.
+- Bundle hosts without hooks (Antigravity, Copilot) → use `--print` and
+  copy-paste.
+
+## Fallback mode — summarize the LIVE conversation (`--print`)
+
+On `/agent-handoff --print` (or when no session log exists), summarize the
+current conversation in-model — the original flow. Collect branch
+(`git branch --show-current`), uncommitted changes (`git status --short`),
+recent commits (`git log --oneline -5`), the active roadmap under
+`agents/roadmaps/`, and the task list only when explicitly asked
+("/agent-handoff with tasks"). Then emit:
 
 ```
 ---
@@ -80,12 +119,39 @@ Roadmap: {roadmap file if active, or "none"}
 - {list of files that were edited or are important for context}
 ```
 
+The CLI generator emits the same section set (minus the live-only
+*Repeatable workflow* / *Feedback history* refinements) — the template above
+stays the single source of the section contract.
+
 **Verbatim-first is the load-bearing rule:** the *User instructions* and
 *Feedback history* sections are preserved word-for-word and never compressed —
 lossy re-summarization of the user's own constraints is the exact failure this
 template prevents. Everything else (Done / Key decisions) stays concise.
 
-### 2b. File-artifact mode — `HANDOFF.md` (optional, host-neutral)
+### Present the fallback to the user
+
+Show the handoff prompt in a fenced code block and say:
+
+```
+> Copy this into a new chat to continue where we left off.
+```
+
+## Iron Law — printed output is ALWAYS a fenced code block
+
+```
+THE HANDOFF MUST BE WRAPPED IN A FENCED ```markdown CODE BLOCK
+SO THE USER CAN COPY IT IN ONE CLICK. NEVER RENDER IT AS LIVE
+MARKDOWN. NEVER SPLIT IT ACROSS MULTIPLE BLOCKS.
+```
+
+Applies to every in-chat emission (`--print` fallback, bundle hosts). Use
+` ```markdown ` as the fence so consumers get syntax highlighting and the
+inner content is preserved verbatim. Add one short prose line above the
+block (e.g. *"Copy this into a new chat:"*) and nothing after — no follow-up
+questions, no numbered options. The picker in Step 2 is the one exception:
+it is the flow's single question, asked BEFORE any handoff is generated.
+
+## 2b. File-artifact mode — `HANDOFF.md` (optional, host-neutral)
 
 On `/agent-handoff --file` (or when a workflow skill's phase boundary asks
 for a standing handoff), ALSO write the contract to
@@ -119,27 +185,6 @@ next to it (`HANDOFF.md.<ts>.bak`) before overwrite; duplicate-check before
 appending (never double-append a section); preserve the section structure;
 post-verify the write by re-reading the required fields.
 
-### 3. Present to user
-
-Show the handoff prompt in a fenced code block and say:
-
-```
-> Copy this into a new chat to continue where we left off.
-```
-
-## Iron Law — output is ALWAYS a fenced code block
-
-```
-THE HANDOFF MUST BE WRAPPED IN A FENCED ```markdown CODE BLOCK
-SO THE USER CAN COPY IT IN ONE CLICK. NEVER RENDER IT AS LIVE
-MARKDOWN. NEVER SPLIT IT ACROSS MULTIPLE BLOCKS.
-```
-
-Use ` ```markdown ` as the fence so consumers get syntax highlighting
-and the inner content is preserved verbatim. Add one short prose line
-above the block (e.g. *"Copy this into a new chat:"*) and nothing
-after — no follow-up questions, no numbered options.
-
 ## Detection — when natural-language triggers count as explicit
 
 The user does not need to type `/agent-handoff` literally. Treat as
@@ -149,6 +194,9 @@ explicit invocation when the prompt contains a verbatim mention of:
 - `fresh chat`, `neuer chat`, `for a new chat`
 - `context summary for {fresh,new} chat`
 - `fasse … für … chat zusammen`
+- `resume {a,the} previous session`, `continue {an,the} old session`
+- `neue session mit dem stand von {gestern,letzter session}`,
+  `alte session … weitermachen`
 
 A vague *"summarize this conversation"* without handoff/new-chat
 framing → NOT a trigger. Surface `/agent-handoff` as a numbered option,
@@ -156,6 +204,9 @@ do not auto-execute.
 
 ## Rules
 
+- **Primary flow is CLI-backed.** When `agent-config` is reachable, route
+  through the picker (Steps 1–4); fall back to the in-model summary only on
+  `--print`, on bundle hosts, or when the CLI reports no sessions.
 - **Concise everywhere EXCEPT the verbatim sections.** Keep Done / Key
   decisions tight, but never truncate *User instructions* or *Feedback history*
   to hit a line target — losing a user constraint costs far more than the
@@ -168,13 +219,13 @@ do not auto-execute.
 - **Open tasks are critical** — the new chat needs to know what's left.
 - **Decisions are important** — prevents the new chat from re-asking settled questions.
 - **File list is optional** — only include if the new chat will need to edit specific files.
-- **NEVER render the handoff as live markdown** — see Iron Law above.
+- **NEVER render a printed handoff as live markdown** — see Iron Law above.
 
 ## When to use this vs. `agents/runtime/.agent-chat-history`
 
-- `/agent-handoff` is **push-based**: you copy a short summary into the
-  new chat. Works across tools (Augment → Claude Code), across machines,
-  and without any persistent file.
+- `/agent-handoff` picks a **past session** and seeds the next one (or, with
+  `--print`, pushes a copy-paste summary of the live one). Works across
+  tools; the auto-inject path needs a hook-capable host on the same machine.
 - `agents/runtime/.agent-chat-history` is **pull-based** and **multi-session**: every
   session writes its own entries tagged with a 16-char session
   fingerprint derived from the platform `session_id` (schema v4, see
@@ -185,13 +236,14 @@ do not auto-execute.
   session for project-improving learnings with
   `/memory mine-session --mode=proposals`.
 
-Prefer `/agent-handoff` for planned context switches across tools or
-machines; use `/chat-history import` after a crash or fresh-chat reopen
-on the same workspace to surface prior-session context verbatim.
+Prefer `/agent-handoff` for planned context switches and session resumes;
+use `/chat-history import` after a crash or fresh-chat reopen on the same
+workspace to surface prior-session context verbatim.
 
 Three distinct mechanisms — do not conflate them:
 
-- **handoff** (this command) — a one-shot push summary for the *next* chat;
-  ephemeral, copy-paste, verbatim on the user's instructions.
+- **handoff** (this command) — a one-shot seed for the *next* chat;
+  generated from a picked session (or pushed from the live one), ephemeral,
+  verbatim on the user's instructions.
 - **[`chat-history import`](../chat-history/import/command.md)** — pull a prior *session's* logged context into the current chat.
 - **durable memory** ([`memory-consolidation`](../../../skills/memory-consolidation/SKILL.md)) — cross-*run* curated facts; a handoff is not memory, and memory is not a transcript.

package/dist/agent-src/commands/commit.md

@@ -193,7 +193,7 @@ Read `verbosity.post_action_reports` from `.agent-settings.yml` (default
 
 ## Rules
 
-- **Never commit before the user confirms** the plan.
+- **Preview path: never commit before the user confirms the plan — once, for the whole plan.** On the terse path (`preview_artifacts: false` AND `routine_confirmations: false`) the `/commit` invocation itself is the confirmation (`commit-policy` exception 3); proceed without a prompt.
 - **Never push** — pushing is left to the user.
 - **Never modify files** — only stage and commit existing changes.
 - **Do NOT add untracked files** unless they are clearly part of the change (check with `git status`).

package/dist/agent-src/commands/condense.md

@@ -20,8 +20,8 @@ packs:
 
 # condense
 
-Condense agent config `.md` files from `src/` into token-efficient telegraph format
-and write the condensed output to `dist/agent-src/`.
+Project agent config `.md` files from `src/` into `dist/agent-src/` — verbatim
+copy with the path rewriter applied (ADR-201; the LLM prose rewrite is removed).
 
 The projection is deterministic: `dist == rewrite(src)` byte-for-byte. A file is
 out of date only when its projection does not match that rewrite.
@@ -96,76 +96,31 @@ Read `verbosity.post_action_reports` from `.agent-settings.yml` (default
 `minimal`).
 
 - `off` → emit nothing on success; surface errors only.
-- `minimal` (default) → one line: `→ N files condensed (avg X% savings)`.
-- `full` → multi-line table with per-category stats (files condensed,
-  avg savings).
+- `minimal` (default) → one line: `→ N files synced`.
+- `full` → multi-line table with per-category stats (files synced,
+  stale files reaped).
 
-## Iron Laws — do not touch
+## Iron Laws — preserved by construction
 
 Sections under headings matching `Iron Law`, `Iron Laws`, or `The Iron Law` (any
 heading level, numbered variants like `Iron Law 1` included) are **load-bearing
-behavioral rules**. Condensation rules above do **not** apply to them.
-
-For every Iron Law section in a source file:
-
-- **Copy the heading verbatim**, exact text, exact `#` level. NEVER downgrade
-  `## Iron Law` to `### Iron Law` or to inline `**Iron Law:**`.
-- **Copy the fenced code block byte-for-byte**, including capitalization, line
-  breaks, and trailing punctuation.
-- **Copy the negation clauses verbatim** — `NO X`, `NEVER Y`, `NOT Z`. These
-  are the law's exception denials; stripping them weakens the rule.
-- **Telegraph the prose, keep every passage** — every paragraph, every list
-  item, and every fenced code block from the source must appear in the
-  condensed output, in order. Drop articles, shorten phrasing, primitive
-  grammar, terse cave-speak — all encouraged. What's forbidden is dropping
-  whole sentences, merging two paragraphs into one, or skipping a bullet.
-  One paragraph → one paragraph; one bullet → one bullet.
-- **No word-count budget** — condense the prose as hard as telegraph style
-  allows. The check is structural (passage count), not quantitative.
-
-`scripts/check_condensation.ts` enforces these mechanically — `iron_law_missing`,
-`iron_law_passage_dropped`, and `iron_law_heading_downgrade` are `error`-level
-and block CI.
+behavioral rules**. Post-ADR-201 the projection is a verbatim copy, so headings,
+fenced blocks, and negation clauses survive by construction — there is no manual
+preservation step for the agent to perform.
 
-If an Iron Law section genuinely contains filler (rare): edit the SOURCE in
-`src/`, not the condensed copy. Source is the truth.
-
-## Condensation quality checklist
-
-**Also apply the [preservation-guard](../rules/preservation-guard.md) rule** — strongest validation, example, anti-pattern, and decision hints must survive condensation. Iron Laws are non-negotiable.
-
-See also: [markdown-safe-codeblocks](../rules/markdown-safe-codeblocks.md) for fenced-block hygiene.
-
-After condensing each file, verify:
+`scripts/check_condensation.ts` enforces this mechanically — since ADR-201 it
+asserts `dist == rewrite(src)` byte-for-byte; `iron_law_missing`,
+`iron_law_passage_dropped`, and `iron_law_heading_downgrade` remain `error`-level
+diagnostics that name what differs when the bytes do.
 
-- [ ] All code blocks preserved exactly (no content changes)
-- [ ] All inline code, URLs, file paths unchanged
-- [ ] YAML frontmatter identical to source
-- [ ] Headings match source exactly
-- [ ] Tables structure preserved (cell text may be shortened)
-- [ ] "NEVER", "MUST", "Do NOT" and other strong language preserved
-- [ ] Technical terms, library names, API names unchanged
-- [ ] No meaning lost — condensed version says the same thing, shorter
-- [ ] No sections accidentally removed
-- [ ] Word count reduction is 20-50% (typical range for prose-heavy files)
-
-### Safe vs unsafe condensation
-
-Safe:
-- "You should always make sure to run the tests before committing" → "Run tests before commit"
-- "In order to ensure that the configuration is correct" → "To verify config"
-- "It is important to note that this feature requires" → "Requires"
-
-Unsafe (DO NOT do this):
-- Removing a bullet point that contains unique information
-- Changing `php artisan test --filter=MyTest` to `php artisan test`
-- Shortening "Do NOT use float for money" to "Avoid float"
-- Removing "NEVER" from "NEVER add to phpstan-baseline.neon"
+If an Iron Law section genuinely contains filler (rare): edit the SOURCE in
+`src/`, not the projected copy. Source is the truth. Authoring-time transforms
+(merges, splits, refactors) remain governed by
+[preservation-guard](../rules/preservation-guard.md).
 
 ## Rules
 
 - **Do NOT commit or push.** Only write files.
 - **Do NOT modify `src/`** — it is the source of truth.
-- **Only write to `.augment/`** — the condensed output directory.
-- **Preserve ALL technical content** — only condense natural language prose.
-- **YAML frontmatter** in command/skill files must be preserved exactly.
+- **Only write to `dist/agent-src/`** (via `condense.sh` / the script) — never
+  hand-edit any projection.

package/dist/agent-src/commands/evals/agent-handoff.json

@@ -1,6 +1,6 @@
 {
   "command": "agent-handoff",
-  "intent": "Generate a session-handoff summary to continue work in a fresh chat",
+  "intent": "Resume a previous session in a fresh chat: pick a session, generate a handoff, auto-seed the next session",
   "cases": [
     {
       "prompt": "summarize where we are so I can paste it into a new chat",
@@ -29,6 +29,18 @@
     {
       "prompt": "show me the current message count and token cost",
       "expected": "agent-status"
+    },
+    {
+      "prompt": "let me continue an old session in a fresh chat",
+      "expected": "agent-handoff"
+    },
+    {
+      "prompt": "starte eine neue session mit dem stand von gestern",
+      "expected": "agent-handoff"
+    },
+    {
+      "prompt": "resume my previous session where we left off, but in a clean chat",
+      "expected": "agent-handoff"
     }
   ]
 }

package/dist/agent-src/commands/memory/promote.md

@@ -184,7 +184,20 @@ echo '{"type":"supersede","supersedes":"<sig-id>","promoted_to":"<curated id>"}'
 This keeps the intake stream truthful — lookups will ignore the
 superseded signal.
 
-### 6. Open a promotion PR
+### 6. Open a promotion PR — git ops are permission-gated
+
+Branch, commit, and push are NOT authorized by invoking this command
+(per `scope-control` § git-ops and the `non-destructive-by-default`
+push floor — invoking `/memory:promote` names the promotion, not a git
+destination). Surface the exact operations and wait for explicit
+confirmation this turn:
+
+> Promotion is written. Ship it as a PR?
+>
+> 1. Yes — branch `memory/promote-<curated-id>`, commit, push, then `/create-pr`
+> 2. No — keep the promotion local, I'll ship it myself
+
+Only on option 1:
 
 ```bash
 git checkout -b memory/promote-<curated-id>

package/dist/agent-src/commands/mode.md

@@ -63,7 +63,9 @@ Extract `roles.default_role` and `roles.active_role`.
 
 Echo the full contract body for the target mode from `role-contracts.md`:
 
-- Goal / Constraints / Output fields
+- The mode's contract-skeleton fields (each mode's skeleton differs —
+  e.g. developer: Goal / Plan / Changes / Tests / Open questions;
+  planner: Goal / Constraints / Option set / …)
 - Default skills the mode loads first
 - Structured mode marker the agent will emit when closing
 
@@ -85,7 +87,7 @@ Do NOT write anything for `/mode` without an argument (status only).
 Announce the switch as:
 
 ```
-> entering {mode} mode — contract: {goal-field} / {constraints-field} / {output-field}
+> entering {mode} mode — contract fields: {first-3-field-names-from-skeleton}
 ```
 
 From this point in the session: the rule
@@ -105,7 +107,7 @@ numbered prompt:
 ## Output format
 
 ```
-> entering {mode} mode — contract: …
+> entering {mode} mode — contract fields: …
 >
 > Default skills: {skill-a}, {skill-b}, {skill-c}
 > Closing marker: <!-- role-mode: {mode} | contract: … -->
@@ -121,8 +123,10 @@ For `/mode` (status only):
 ## Gotchas
 
 - `.agent-settings.yml` is git-ignored. `/mode` never commits the file.
-- If the file is missing the `roles:` section, add it in the correct
-  place (between `subagents:` and EOF) using the template layout.
+- If the file is missing the `roles:` section, add it — note
+  `agent-settings.template.yml` ships no `roles:` block; the reference
+  `roles:` layout lives in
+  `src/agent-src/templates/agents/agent-project-settings.example.yml`.
 - Mode names are case-sensitive in the file; case-insensitive on input.
 - The contract is the source of truth, not this command — if the
   guideline changes, this command reflects the new contract on next run.