@event4u/agent-config 5.10.1 → 6.1.0

package/docs/contracts/smoke-contracts.md

@@ -3,7 +3,7 @@ stability: beta
 keep-beta-until: 2026-08-14
 ---
 
-# Smoke Contracts — Phase 3 of step-11-ruflo-parity
+# Smoke Contracts — Phase 3 of an internal parity roadmap (local-only)
 
 > **Status:** active · **Owner:** step-11 Phase 3 · **Sibling:**
 > [`measurement-baseline.md`](measurement-baseline.md) (snapshot semantics)
@@ -34,7 +34,7 @@ the paths touched in the PR:
 
 | Tier | Globs that trigger | Script |
 |---|---|---|
-| kernel | `.agent-src.uncondensed/rules/**`, `.agent-src/rules/**`, `dist/router.json`, `scripts/measure_rule_budget.py` | `scripts/smoke/kernel.sh` |
+| kernel | `.agent-src.uncondensed/rules/**`, `dist/agent-src/rules/**`, `dist/router.json`, `scripts/measure_rule_budget.py` | `scripts/smoke/kernel.sh` |
 | router | `dist/router.json`, `.agent-src.uncondensed/rules/**`, `.agent-src.uncondensed/skills/**`, `docs/contracts/**`, `docs/guidelines/**` | `scripts/smoke/router.sh` |
 | schema | `.agent-src.uncondensed/skills/**`, `.agent-src.uncondensed/rules/**`, `scripts/schemas/**`, `scripts/skill_linter.py`, `scripts/validate_frontmatter.py` | `scripts/smoke/schema.sh` |
 | skills | `.agent-src.uncondensed/skills/**` | `scripts/smoke/skills.sh` |
@@ -74,7 +74,7 @@ constant in the script body and the row below.
 ```
 
 - **68 ids** — 10 kernel + 23 tier_1 + 35 tier_2; every id resolves to
-  `.agent-src/rules/<id>.md`.
+  `dist/agent-src/rules/<id>.md`.
 - **0 broken rule pointers** — hard assertion; smoke fails on any miss.
 - **36 routes_to refs** across tier_1 + tier_2; resolver honours the
   four prefixes (`skill:`, `command:`, `guideline:`, `contract:`).
@@ -110,7 +110,7 @@ constant in the script body and the row below.
   `.agent-src.uncondensed/skills/*/SKILL.md` and re-validated via
   `scripts/validate_frontmatter.py`. `agent-config explain skill` is
   **not** invoked — `explain` only supports `{config,rule,route}` today
-  ([`scripts/agent-config/cmd_explain.py`](../../scripts/agent-config/cmd_explain.py));
+  ([`scripts/agent-config/cmd_explain.py`](../../src/scripts/agent-config/cmd_explain.py));
   filesystem-resolution is the contract.
 
 ## § 4 — Local invocation
@@ -132,7 +132,7 @@ the final baseline line) for CI summary parsing.
 |---|---|---|
 | `kernel.sh` reports > 9 missing fences | Kernel rule lost its Iron Law block during edit | Restore the fence; update `EXEMPT_FROM_FENCE` only for new dispatch indexes |
 | `router.sh` reports > 0 broken pointers | `dist/router.json` references an id without a rule file | Add the rule or remove the route — never edit the smoke baseline up |
-| `schema.sh` reports FAILs | A skill / rule lost a required field | Restore via [`scripts/schemas/skill.schema.json`](../../scripts/schemas/skill.schema.json) |
+| `schema.sh` reports FAILs | A skill / rule lost a required field | Restore via [`scripts/schemas/skill.schema.json`](../../src/scripts/schemas/skill.schema.json) |
 | `skills.sh` 5/5 random sample fails | Hand-edit broke frontmatter or renamed directory without updating `name:` | Restore filename ↔ slug coupling |
 
 ## § 6 — See also

package/docs/contracts/telegraph-telemetry.md

@@ -66,10 +66,10 @@ Until a v2 bench (broader corpus or a re-tuned dialect) lifts the
 
 ## Consumers
 
-- [`scripts/telegraph_stats.py`](../../scripts/telegraph_stats.py) — reads
+- [`scripts/telegraph_stats.py`](../../src/scripts/telegraph_stats.py) — reads
   this constant, computes per-session / per-conversation / lifetime
   deltas from `agents/cost-tracking/sessions.jsonl`.
-- [`scripts/cost_summary.py`](../../scripts/cost_summary.py) — emits
+- [`scripts/cost_summary.py`](../../src/scripts/cost_summary.py) — emits
   the stable JSON contract for inter-tool consumption per
   [`cost-summary-schema.md`](cost-summary-schema.md).
 - `agent-status` skill — surfaces the per-session delta in the

package/docs/contracts/trust-and-safety.md

@@ -109,7 +109,7 @@ asserts three invariants over `dist/discovery/discovery-manifest.json`:
    least one artefact whose logical path contains `safety-floor`.
 2. **Banner presence.** Every artefact with
    `trust.human_review_required: true` has its compiled output under
-   `.agent-src/<logical>` and that output contains
+   `dist/agent-src/<logical>` and that output contains
    `<!-- agent-config:human-review-banner -->`.
 3. **Kernel trust floor.** Every rule listed in `dist/router.json` `kernel[]`
    declares `trust.level: core`. No escalation (advisory blocks the
@@ -135,10 +135,10 @@ Exits 0 clean, 1 on any violation. Tests in
 ## § 7 — References
 
 - ADR: [`ADR-018 — Trust & Safety Layer`](../decisions/ADR-018-trust-and-safety-layer.md)
-- Condenseor implementation: [`scripts/condense.py`](../../scripts/condense.py)
+- Condenseor implementation: [`scripts/condense.py`](../../src/scripts/condense.py)
   (`_inject_hrr_banner`, `_HRR_BANNER_MARKER`)
 - Installer implementation: [`packages/core/installer/src/trust-escalation.ts`](../../packages/core/installer/src/trust-escalation.ts)
-- Lint implementation: [`scripts/lint_trust_coherence.py`](../../scripts/lint_trust_coherence.py)
+- Lint implementation: [`scripts/lint_trust_coherence.py`](../../src/scripts/lint_trust_coherence.py)
 - Sibling contracts: [`safety-model`](safety-model.md) ·
   [`kernel-membership`](kernel-membership.md) ·
   [`ADR-013` — discovery frontmatter](../decisions/ADR-013-discovery-frontmatter-contract.md)

package/docs/contracts/universal-skills.md

@@ -18,21 +18,21 @@ The allowlist solves the over-siloing risk identified by the AI Council 2026-05-
 
 | Skill | Why universal |
 |---|---|
-| [`refine-prompt`](../../.agent-src/skills/refine-prompt/SKILL.md) | Every free-form request needs AC + assumption + confidence band before the engine runs. |
-| [`refine-ticket`](../../.agent-src/skills/refine-ticket/SKILL.md) | Every ticket (Jira / Linear / GH issue) needs scoped AC before plan. |
-| [`estimate-ticket`](../../.agent-src/skills/estimate-ticket/SKILL.md) | Sizing + split + risk — applies to any work item, technical or not. |
-| [`verify-completion-evidence`](../../.agent-src/skills/verify-completion-evidence/SKILL.md) | Iron-Law gate: fresh evidence before any "done" claim. Survives every mode. |
-| [`threat-modeling`](../../.agent-src/skills/threat-modeling/SKILL.md) | Pre-implementation abuse-case enumeration — same lens applies to a marketing form and a payment endpoint. |
-| [`systematic-debugging`](../../.agent-src/skills/systematic-debugging/SKILL.md) | Reproduce → isolate → hypothesize → verify — works on broken pipelines, broken docs, broken funnels. |
-| [`doc-coauthoring`](../../.agent-src/skills/doc-coauthoring/SKILL.md) | 3-stage write loop (context → section → reader-test) for any long-form doc. |
-| [`deep-reading-analyst`](../../.agent-src/skills/deep-reading-analyst/SKILL.md) | Article / long-form analysis via SCQA + mental models — used by every role that reads. |
-| [`decision-record`](../../.agent-src/skills/decision-record/SKILL.md) | Trade-off framing before any commit. ADR-shape works for tech, hiring, and pricing. |
-| [`adr-create`](../../.agent-src/skills/adr-create/SKILL.md) | Naming + numbering + index regen — the file mechanics of `decision-record`. |
-| [`risk-officer`](../../.agent-src/skills/risk-officer/SKILL.md) | Blast-radius framing + residual-risk verdict before any commit. |
-| [`adversarial-review`](../../.agent-src/skills/adversarial-review/SKILL.md) | Devil's advocate stress-test — opt-in but always available. |
-| [`customer-research`](../../.agent-src/skills/customer-research/SKILL.md) | Discovery-slice shaping (JTBD, switch-event) — used by founder, GTM, consultant, creator. |
-| [`stakeholder-tradeoff`](../../.agent-src/skills/stakeholder-tradeoff/SKILL.md) | Per-lens framing when stakeholders disagree — applies in every org. |
-| [`md-language-check`](../../.agent-src/skills/md-language-check/SKILL.md) | Hard gate: `.md` files stay English; the agent translates at runtime. |
+| [`refine-prompt`](../../dist/agent-src/skills/refine-prompt/SKILL.md) | Every free-form request needs AC + assumption + confidence band before the engine runs. |
+| [`refine-ticket`](../../dist/agent-src/skills/refine-ticket/SKILL.md) | Every ticket (Jira / Linear / GH issue) needs scoped AC before plan. |
+| [`estimate-ticket`](../../dist/agent-src/skills/estimate-ticket/SKILL.md) | Sizing + split + risk — applies to any work item, technical or not. |
+| [`verify-completion-evidence`](../../dist/agent-src/skills/verify-completion-evidence/SKILL.md) | Iron-Law gate: fresh evidence before any "done" claim. Survives every mode. |
+| [`threat-modeling`](../../dist/agent-src/skills/threat-modeling/SKILL.md) | Pre-implementation abuse-case enumeration — same lens applies to a marketing form and a payment endpoint. |
+| [`systematic-debugging`](../../dist/agent-src/skills/systematic-debugging/SKILL.md) | Reproduce → isolate → hypothesize → verify — works on broken pipelines, broken docs, broken funnels. |
+| [`doc-coauthoring`](../../dist/agent-src/skills/doc-coauthoring/SKILL.md) | 3-stage write loop (context → section → reader-test) for any long-form doc. |
+| [`deep-reading-analyst`](../../dist/agent-src/skills/deep-reading-analyst/SKILL.md) | Article / long-form analysis via SCQA + mental models — used by every role that reads. |
+| [`decision-record`](../../dist/agent-src/skills/decision-record/SKILL.md) | Trade-off framing before any commit. ADR-shape works for tech, hiring, and pricing. |
+| [`adr-create`](../../dist/agent-src/skills/adr-create/SKILL.md) | Naming + numbering + index regen — the file mechanics of `decision-record`. |
+| [`risk-officer`](../../dist/agent-src/skills/risk-officer/SKILL.md) | Blast-radius framing + residual-risk verdict before any commit. |
+| [`adversarial-review`](../../dist/agent-src/skills/adversarial-review/SKILL.md) | Devil's advocate stress-test — opt-in but always available. |
+| [`customer-research`](../../dist/agent-src/skills/customer-research/SKILL.md) | Discovery-slice shaping (JTBD, switch-event) — used by founder, GTM, consultant, creator. |
+| [`stakeholder-tradeoff`](../../dist/agent-src/skills/stakeholder-tradeoff/SKILL.md) | Per-lens framing when stakeholders disagree — applies in every org. |
+| [`md-language-check`](../../dist/agent-src/skills/md-language-check/SKILL.md) | Hard gate: `.md` files stay English; the agent translates at runtime. |
 
 ## Inclusion criteria
 
@@ -44,7 +44,7 @@ A skill earns the universal label when **all three** hold:
 
 ## Exclusion criteria
 
-- **Domain anchor.** `voice-and-tone-design` (creator), `runway-cognition` (founder), `quality-tools` (developer), `pipeline-strategy` (GTM) — strong-fit for one user-type, weak for the other five. These tag via the `recommended_for_user_types:` frontmatter field defined in [`skill.schema.json`](../../scripts/schemas/skill.schema.json).
+- **Domain anchor.** `voice-and-tone-design` (creator), `runway-cognition` (founder), `quality-tools` (developer), `pipeline-strategy` (GTM) — strong-fit for one user-type, weak for the other five. These tag via the `recommended_for_user_types:` frontmatter field defined in [`skill.schema.json`](../../src/scripts/schemas/skill.schema.json).
 - **Tool-specific.** `laravel-horizon`, `nextjs-patterns`, `mcp-builder`, `terraform` — load only when the matching stack flag is set.
 - **Council / role review.** `judge-bug-hunter`, `judge-security-auditor`, `architecture-review-lens` — load via subagent orchestration, not via universal allowlist.
 
@@ -87,6 +87,5 @@ Removing a skill is breaking. It must be announced in `STABILITY.md` with a depr
 ## Related
 
 - [`STABILITY.md`](STABILITY.md) — public-identifier stability surface.
-- [`agent-memory-contract.md`](agent-memory-contract.md) — companion contract for the cross-cutting memory layer.
 - [`docs/getting-started-by-role.md`](../getting-started-by-role.md) — consumer-facing role docs that reference these skills.
 - [`tests/eval/corpus-non-dev.yaml`](../../tests/eval/corpus-non-dev.yaml) — empirical test corpus.

package/docs/contracts/value-dashboard-spec.md

@@ -24,9 +24,9 @@ machine-readable source of truth.
 ## Source
 
 - **Chat thread:** 2026-05-27 (the owner's verdict: *"Aktuell bringen
-  diese Benchmarks nichts. Ich weiß worum es geht und verstehe sie nicht
+  diese Benchmarks nichts. Ich weiß worum es geht und verstehe sie nicht  <!-- md-language-check: ignore -->
   mal."*)
-- **Roadmap:** `agents/roadmaps/road-to-readable-value-dashboard.md`
+- **Roadmap:** `road-to-readable-value-dashboard` (archived)
 - **Extends:** archived `road-to-package-impact-benchmark.md` (A/B
   surface) and archived `step-4-measurement-and-benchmark.md`
   (telegraph + selection bench).
@@ -70,18 +70,18 @@ Each rung is a measurement (or a clearly-marked `pending` placeholder),
 not a marketing number. The ladder is **honest about the up-front
 cost**: installing the package first *adds* input tokens (rules load
 into context); condense + rtk + terse then claw that back. The
-**NETTO** line is the real answer.
+**NET** line is the real answer.
 
 Rung order:
 
-1. **Ohne Paket** — baseline. Token delta = 0. Reference rung; the
+1. **Without package** — baseline. Token delta = 0. Reference rung; the
    ladder is computed relative to this.
-2. **Mit Paket (Regeln laden)** — the honest up-front cost. The
+2. **With package (rule load)** — the honest up-front cost. The
    always-loaded kernel + router footprint added to every request's
    input. Token delta is *positive* (the rule body lands in context).
    Source: `metric_a_footprint` from `frugality/baseline.jsonl`.
 3. **+ condense** — the input-side carve-out savings from
-   condensation (`.agent-src.uncondensed` → `.agent-src`). Source:
+   condensation (`.agent-src.uncondensed` → `dist/agent-src`). Source:
    `internal/bench/reports/telegraph-v2.json`. **Excludes Thin-Root
    files** (AGENTS.md variants — they net negative); the Thin-Root
    caveat surfaces as a footnote, not a hidden exclusion.
@@ -100,7 +100,7 @@ Rung order:
 
 The renderer prints, per rung:
 
-- **label** — short German + English ("Mit Paket / +load")
+- **label** — short English ("With package (rule load) / +load")
 - **what-it-does** — one phrase a non-developer understands
 - **token_delta** — signed integer (positive = adds tokens)
 - **eur_delta** — token_delta priced at the reference scale below
@@ -144,7 +144,7 @@ Every rung in the `cost_ladder` array carries:
 
 ```yaml
 id: <kebab-case>                  # e.g. "load", "condense", "rtk"
-label: "<German + English>"       # e.g. "Mit Paket (Regeln laden)"
+label: "<English>"                # e.g. "With package (rule load)"
 what_it_does: "<one phrase>"      # plain language, ≤ 80 chars
 token_delta: <signed int>         # per-request input token delta
 eur_delta: <float>                # token_delta priced at reference scale
@@ -212,7 +212,7 @@ Each metric in the `behaviour` block carries:
 
 ```yaml
 id: <kebab-case>                  # e.g. "selection", "destructive-stops"
-label: "<German + English>"
+label: "<English>"
 what_this_means: "<one line>"     # plain language caption
 with: <value>                     # metric-specific (pct, count, ratio)
 without: <value>
@@ -237,9 +237,9 @@ corpus. This roadmap considered two options:
 product (per the non-goals section of the roadmap). Hiding the
 negative number — even by relocating it to a "quality" panel —
 would betray the for-dummies-honest framing. The rung renders with
-its measured value, `confidence: measured`, and a caption: *"In
-unserem Testkorpus liefert Telegraph mehr Tokens als ein neutrales
-'sei knapp' — wir messen, wir verstecken nicht."*
+its measured value, `confidence: measured`, and a caption: *"In our
+test corpus, telegraph delivers more tokens than a neutral
+'be terse' — we measure, we do not hide."*
 
 ## Glossary (rendered into `docs/value.md`)
 
@@ -258,15 +258,15 @@ copies it verbatim into the dashboard.
   than input. Per-token output costs more than input.
 - **condense** — a build step that shrinks the rule files
   before they ship (`.agent-src.uncondensed` →
-  `.agent-src`). Saves input tokens on every request.
+  `dist/agent-src`). Saves input tokens on every request.
 - **rtk** — the *Rust Token Killer*, a CLI wrapper that strips
   verbose output (`git status`, lint output, test runners) before
   the model reads it. Saves input tokens on tool calls.
 - **terse / telegraph** — a style of output (short phrases,
   dropped articles) the agent uses when condensing replies.
   Saves output tokens — when the corpus rewards it.
-- **Ohne Paket / Mit Paket** — "without the package" /
-  "with the package" — the two arms of the A/B comparison.
+- **Without package / With package** — the two arms of the A/B
+  comparison.
 - **Δ Tokens** — input-token difference per request vs. the baseline.
   The rendered dashboard reports cost in **tokens only** — no € figure.
   A €/USD comparison would assume per-call API pricing, which the many
@@ -292,9 +292,9 @@ actual always-loaded kernel. The real kernel has 10 rules per
 | Metric | Before fix | After fix | Delta |
 |---|---:|---:|---:|
 | `load` token delta | +4 843 | **+8 977** | +4 134 |
-| NETTO token delta | +4 120 | **+8 254** | +4 134 |
-| NETTO `cumulative_pct` | +51.5 % | **+103.2 %** | +51.7 pp |
-| NETTO €/1k requests | +€11.37 | **+€22.78** | +€11.41 |
+| NET token delta | +4 120 | **+8 254** | +4 134 |
+| NET `cumulative_pct` | +51.5 % | **+103.2 %** | +51.7 pp |
+| NET €/1k requests | +€11.37 | **+€22.78** | +€11.41 |
 
 The original dashboard under-reported the base-load by ~4 100
 tokens/request. Panel B's behaviour numbers are unaffected (they
@@ -309,7 +309,7 @@ measure agent behaviour, not token footprint).
 - Phase 5 — tier-1 dead-rule audit closed with zero cuts: of 20 never-matched rules, 19 are bench-blind / measurement-window / cluster-head (load-bearing despite zero corpus hits); the lone demote candidate (`symfony-routing`) is kept to preserve cross-stack portability.
 - Phase 6 — full live Track B re-run skipped: Phase 1-5 made no rule-body or frontmatter edits, so by construction Panel B is unchanged from the 2026-05-28 baseline (`with` 84.6 % completion, `without` 7.7 % completion). Re-running would consume tokens to re-confirm a known value.
 
-**Pass outcome:** NETTO moved from +4 120 (mis-measured) to **+8 254 tokens / request** (honest); Panel B held by construction. The pass's value is the corrected measurement floor + the new telemetry tooling, not any in-place rule cuts. Cuts must wait until the bench corpus is widened to exercise the rules' real trigger surfaces (git, onboarding, roadmap work, long-conversation windows, autonomy moments).
+**Pass outcome:** NET moved from +4 120 (mis-measured) to **+8 254 tokens / request** (honest); Panel B held by construction. The pass's value is the corrected measurement floor + the new telemetry tooling, not any in-place rule cuts. Cuts must wait until the bench corpus is widened to exercise the rules' real trigger surfaces (git, onboarding, roadmap work, long-conversation windows, autonomy moments).
 
 **Optimisation pass 2 close-out (2026-05-29, `road-to-corpus-expansion-evidence-based-cuts`):**
 
@@ -320,7 +320,7 @@ measure agent behaviour, not token footprint).
 - Phase 5 — zero cuts (0 demotes, 0 deletes). Same outcome as pass-1, but for a fundamentally different reason: pass-1 closed with zero cuts because the corpus was blind; pass-2 closes with zero cuts because the widened corpus **proved every tier-1 rule has structural reason to exist**.
 - Phase 6 — full live Track B re-run skipped: Phase 1-5 made zero rule-body / frontmatter / kernel edits — Panel B is unchanged from the 2026-05-28 baseline by construction.
 
-**Pass outcome:** NETTO unchanged at **+8 254 tokens / request** (+103.2 % vs. baseline, +€22.78 per 1 000 requests). The pass's actual deliverable is the **structural categorisation** of the 20 previously-never-matched rules — future audits no longer need to re-debate why these rules don't fire in standard corpora. 5 state-bound + 5 intent-only are permanently classified as router-replay-unreachable. The Pass B (kernel-body refactor) deferral remains intact — no candidate qualifies under the tightened thresholds.
+**Pass outcome:** NET unchanged at **+8 254 tokens / request** (+103.2 % vs. baseline, +€22.78 per 1 000 requests). The pass's actual deliverable is the **structural categorisation** of the 20 previously-never-matched rules — future audits no longer need to re-debate why these rules don't fire in standard corpora. 5 state-bound + 5 intent-only are permanently classified as router-replay-unreachable. The Pass B (kernel-body refactor) deferral remains intact — no candidate qualifies under the tightened thresholds.
 
 **Pass B status: deferred / closed for now.** Zero genuine candidates surfaced; the 4 raw-pareto flags are all false-positives. Reopen only when a tier-1 rule both activates frequently in the widened corpus AND has a body that exceeds the kernel-budget ceiling — current state has neither.
 
@@ -370,7 +370,7 @@ deliberately rewriting this section.
 
 ## See also
 
-- [`agents/roadmaps/road-to-readable-value-dashboard.md`](../../agents/roadmaps/road-to-readable-value-dashboard.md) — the roadmap that built this surface.
+- the `road-to-readable-value-dashboard` roadmap (archived) — built this surface.
 - [`value-report-schema.md`](value-report-schema.md) — per-report JSON shape (sibling contract).
 - [`benchmark-ab-contract.md`](benchmark-ab-contract.md) — A/B variant-axis contract (data source for the rtk, behaviour, completion rungs).
 - [`benchmark-report-schema.md`](benchmark-report-schema.md) — per-report JSON shape for A/B reports.

package/docs/contracts/value-report-schema.md

@@ -6,8 +6,8 @@ keep-beta-until: 2026-08-28
 # Value Report Schema (`value-v1`)
 
 Parser-visible contract for the JSON report emitted by
-[`scripts/_lib/value_report.py`](../../scripts/_lib/value_report.py)
-and consumed by [`scripts/render_value_md.py`](../../scripts/render_value_md.py).
+[`scripts/_lib/value_report.py`](../../src/scripts/_lib/value_report.py)
+and consumed by [`scripts/render_value_md.py`](../../src/scripts/render_value_md.py).
 Sibling of [`benchmark-report-schema.md`](benchmark-report-schema.md);
 companion to [`value-dashboard-spec.md`](value-dashboard-spec.md) which
 owns the semantics this contract types.
@@ -102,7 +102,7 @@ notes:
 - **No saving label on negative.** A rung with `token_delta > 0` is a
   *cost* rung; a rung with `token_delta < 0` is a *saving* rung;
   zero is *neutral*. The linter
-  ([`scripts/lint_value_dashboard.py`](../../scripts/lint_value_dashboard.py))
+  ([`scripts/lint_value_dashboard.py`](../../src/scripts/lint_value_dashboard.py))
   rejects any rendered "saving" label on a positive `token_delta`.
 - **No `measured` without a real source.** A rung that carries
   `confidence: measured` MUST have a `source_report` that exists on
@@ -160,9 +160,9 @@ The optional `.md` dump carries:
 ## Cross-references
 
 - Semantics — [`value-dashboard-spec.md`](value-dashboard-spec.md)
-- Roadmap — [`agents/roadmaps/road-to-readable-value-dashboard.md`](../../agents/roadmaps/road-to-readable-value-dashboard.md)
+- Roadmap — the `road-to-readable-value-dashboard` roadmap (archived)
 - Pricing source — [`internal/bench/pricing.yaml`](../../internal/bench/pricing.yaml)
-- Rung normaliser — [`scripts/_lib/value_ladder.py`](../../scripts/_lib/value_ladder.py)
-- Report assembler — [`scripts/_lib/value_report.py`](../../scripts/_lib/value_report.py)
-- Renderer — [`scripts/render_value_md.py`](../../scripts/render_value_md.py)
-- Linter — [`scripts/lint_value_dashboard.py`](../../scripts/lint_value_dashboard.py)
+- Rung normaliser — [`scripts/_lib/value_ladder.py`](../../src/scripts/_lib/value_ladder.py)
+- Report assembler — [`scripts/_lib/value_report.py`](../../src/scripts/_lib/value_report.py)
+- Renderer — [`scripts/render_value_md.py`](../../src/scripts/render_value_md.py)
+- Linter — [`scripts/lint_value_dashboard.py`](../../src/scripts/lint_value_dashboard.py)

package/docs/contracts/workspace-boundary.md

@@ -0,0 +1,84 @@
+---
+stability: stable
+---
+
+# Workspace Boundary Contract
+
+> **Status** · stable · 2026-06-14. Governed by
+> [`ADR-095`](../decisions/ADR-095-workspace-boundary-contract.md). Exists to
+> stop "workspace" from becoming the new "meta" — a catch-all that absorbs
+> every nearby concern until it owns everything and means nothing.
+
+The **workspace** is the task-orchestration layer behind `/work`,
+`/implement-ticket`, and the host-drive loop — 13 modules under
+`src/cli/python/workspace_*.py`. This contract draws the line around it.
+
+## Owns
+
+The workspace owns the **orchestration and lifecycle** of work:
+
+- **Task orchestration** — picking up, sequencing, and driving a task to done.
+- **Host-session lifecycle** — host detection, tier selection, fallback, the
+  drive loop, the kill-switch.
+- **Continuation** — multi-turn resume: a long task survives across turns
+  instead of restarting cold.
+- **Drive health** — task completion / abandonment / retry telemetry, drive
+  kill-switch signals, local workspace analytics **as drive-health data**.
+
+## Does NOT own
+
+The workspace orchestrates these domains but does **not** own their
+*design, semantics, or policy*. It may **consume** their artefacts (read a
+skill body to hand it off; read a profile to launch under it) — it must not
+**define** them:
+
+| Not owned | Owner lives at | The line |
+|---|---|---|
+| **Skill design** | `src/skills/`, `condense.py`, `skill_linter.py` | Workspace may *resolve* a skill body for hand-off; it must not define what a skill is or how it is authored. |
+| **Profile semantics** | `src/profiles/`, `build_discovery_manifest.py` | Workspace may *launch under* a profile; it must not decide what a profile means. |
+| **Video-provider logic** | `src/scripts/ai-video/` | Workspace never reaches into provider adapters or lifecycle. |
+| **MCP-registry policy** | the MCP builder / registry surface | Workspace never encodes which MCP servers exist or how they install. |
+| **Analytics product strategy** | the telemetry / product-analytics surface | Workspace records *drive-health* metrics locally; it does not decide analytics product strategy. |
+
+## Drift detection — import-edge check + doc-governance
+
+Two layers, with an explicit division of labour:
+
+1. **Import-edge linter** (`scripts/lint_workspace_boundary.py`, AST-static,
+   wired into CI). Fails if any `src/cli/python/workspace_*.py` module imports
+   an owner-module of a not-owned domain. This is the cheap, precise lock — it
+   catches the most common concrete drift (a workspace module reaching into
+   video / MCP / skill-design internals).
+
+2. **Doc-governance (review).** The import check enforces **import edges
+   only**. It does **not** catch *semantic* drift — a workspace module that
+   encodes profile-semantics or analytics-*product-strategy* judgement without
+   importing anything forbidden. That stays a **review responsibility** against
+   this contract. **A green import check is not proof the boundary holds** — it
+   is one supplement to boundary thinking, not a substitute for it.
+
+### Escape hatch
+
+A genuinely justified import may carry an inline pragma:
+
+```python
+import some_owner_module  # boundary-exception: <reason>
+```
+
+The linter then allows that line. A `boundary-exception` is reviewed like any
+boundary change — it is a documented, deliberate exception, not a silent
+bypass. Persistent exceptions are a signal to revisit this contract or the
+module's placement.
+
+## Day-one state
+
+Survey of all 13 `workspace_*.py` modules (2026-06-14, recorded in ADR-095):
+**zero violations**. The only cross-module import is intra-workspace
+(`workspace_inbox → workspace_skills`). The import check therefore locks a
+boundary that currently holds.
+
+## See also
+
+- [`ADR-095`](../decisions/ADR-095-workspace-boundary-contract.md) — the decision.
+- [`ADR-050`](../decisions/ADR-050-workspace-vs-package-root-boundary.md) — the workspace-vs-package-root trust boundary (this refines it at module level).
+- [`daily-workspace.md`](daily-workspace.md) — the daily-workspace surface contract.

package/docs/customization.md

@@ -100,7 +100,7 @@ duplicating the root file.
 
 The user-global file is created **only on explicit opt-in via the
 setup wizard** (`agent-config setup`). The loader at
-[`scripts/_lib/agent_settings.py`](../scripts/_lib/agent_settings.py)
+[`scripts/_lib/agent_settings.py`](../src/scripts/_lib/agent_settings.py)
 is **read-only** — no script can create or mutate it without explicit
 wizard confirmation. Edit the file by hand for mid-life changes;
 `/sync-agent-settings` stays project-scoped and never touches
@@ -242,12 +242,12 @@ to `false` in `.agent-settings.yml` to silence post-task analysis without
 changing the profile.
 
 The authoritative matrix of all matrix-controlled settings lives in
-[`.agent-src.uncondensed/templates/agent-settings.md`](../.agent-src.uncondensed/templates/agent-settings.md).
+[`src/agent-src/templates/agent-settings.md`](../src/agent-src/templates/agent-settings.md).
 
 ### Chat-history defaults per profile
 
 `scripts/install.py` fills these placeholders from
-[`config/profiles/*.ini`](../config/profiles) when it writes
+[`src/config/profiles/*.ini`](../config/profiles) when it writes
 `.agent-settings.yml`. Edit the values afterwards if you want different
 behavior — the per-profile table is just the initial default.
 
@@ -280,7 +280,7 @@ output. Iron-Law gates (`commit-policy`, `scope-control` git-ops,
 | `telegraph.speak_scope` | `off`, `prose_only`, `aggressive` | `prose_only` | How widely telegraph-speak grammar applies in chat. `off` = no telegraph; `prose_only` = telegraph in body prose, numbered options + Iron-Law-literal blocks stay full prose; `aggressive` = telegraph everywhere except Iron-Law literals. |
 
 The cross-rule index for these defaults lives in
-[`.agent-src.uncondensed/contexts/contracts/frugality-charter.md`](../.agent-src.uncondensed/contexts/contracts/frugality-charter.md).
+[`src/agent-src/contexts/contracts/frugality-charter.md`](../src/agent-src/contexts/contracts/frugality-charter.md).
 Writer skills (`skill-writing`, `rule-writing`, `command-writing`,
 `guideline-writing`, `roadmap-writing`, `persona-writing`,
 `agent-docs-writing`, `context-authoring`, `conventional-commits-writing`,
@@ -373,7 +373,7 @@ fallback). Files under `~/.event4u/agent-config/agents/settings/contexts/` or
 project-shaped and must not leak across projects.
 
 The resolver lives at
-[`scripts/_lib/agents_overlay.py`](../scripts/_lib/agents_overlay.py)
+[`scripts/_lib/agents_overlay.py`](../src/scripts/_lib/agents_overlay.py)
 and is enforced by `scripts/check_overlay_cascade_subdirs.py` — drift
 between the code constants (`CASCADE_ELIGIBLE_KINDS`,
 `USER_GLOBAL_OVERLAY_KINDS`) and the table above breaks the build.
@@ -387,9 +387,9 @@ discovery keys in its frontmatter — `workspaces:`, `packs:`,
 `lifecycle:`, `trust:`, `install:`. The contract is defined in
 [ADR-013](decisions/ADR-013-discovery-frontmatter-contract.md) and the
 closed vocabularies live in
-[`config/discovery/workspaces.yml`](../config/discovery/workspaces.yml)
+[`src/config/discovery/workspaces.yml`](../config/discovery/workspaces.yml)
 (9 workspaces) and
-[`config/discovery/packs.yml`](../config/discovery/packs.yml) (21
+[`src/config/discovery/packs.yml`](../config/discovery/packs.yml) (21
 packs).
 
 At release time, `scripts/build_discovery_manifest.py` walks the source
@@ -410,7 +410,7 @@ tarball and is the source of truth consumed by:
 > When you read "the Laravel pack", picture a filter over the shared
 > artefact tree — not a separate npm package. The day ADR-011's
 > extraction trigger flips, the same scan output drives the split;
-> until then, every artefact lives under `.agent-src.uncondensed/`.
+> until then, every artefact lives under `src/`.
 
 Alongside the JSON manifest, the scanner writes
 `dist/discovery/discovery-manifest.json.sha256` — a sidecar hash that
@@ -442,7 +442,7 @@ the matching packs' commands/skills are the surfaced set:
 - `/profile deactivate` — full surface returns.
 
 The activatable name is a session-profile alias
-(`config/discovery/session-profiles.yml`: `developer`, `po`, `finance`,
+(`src/config/discovery/session-profiles.yml`: `developer`, `po`, `finance`,
 `gtm`, `content`) or a raw pack id. It is **recommendation-bias only** —
 an inactive-pack command still runs (with a one-line notice); execution is
 never blocked. The choice is **ephemeral**: it writes
@@ -495,8 +495,8 @@ Each project decides; the user-global file cannot flip it on or off
 for unrelated projects.
 
 The decision logic lives at
-[`scripts/_lib/update_check.py`](../scripts/_lib/update_check.py); the
-dispatcher integration lives in [`scripts/agent-config`](../scripts/agent-config)
+[`scripts/_lib/update_check.py`](../src/scripts/_lib/update_check.py); the
+dispatcher integration lives in [`scripts/agent-config`](../src/scripts/agent-config)
 (`run_update_check_banner`).
 
 ---

package/docs/decisions/ADR-006-skill-tools-python-pilot.md

@@ -108,7 +108,7 @@ dir, and the `lint-skill-tools` task entry. Mark this ADR
 - Roadmap: [`road-to-better-skills-and-profiles`](../../agents/roadmaps/archive/road-to-better-skills-and-profiles.md) Block D.
 - Council: [`block-d-python-tools-pilot-verdict`](../../agents/runtime/council/responses/block-d-python-tools-pilot-verdict.md). <!-- council-ref-allowed: ADR decision trace -->
 
-- Tools: [`scripts/skill_tools/`](../../scripts/skill_tools/).
-- Linter: [`scripts/lint_skill_tools.py`](../../scripts/lint_skill_tools.py).
+- Tools: [`scripts/skill_tools/`](../../src/scripts/skill_tools/).
+- Linter: [`scripts/lint_skill_tools.py`](../../src/scripts/lint_skill_tools.py).
 - Corpora + report: [`agents/evidence/eval-corpora/block-d/`](../../agents/evidence/eval-corpora/block-d/).
 - Tests: `tests/test_lint_skill_tools.py`, `tests/test_score_skill_relevance.py`, `tests/test_audit_persona_coverage.py`, `tests/test_suggest_skill_for_task.py`.

package/docs/decisions/ADR-007-agent-discovery-scopes.md

@@ -28,7 +28,7 @@ phase: post-v2.1.0 · simplicity-and-everywhere
 
 **Accepted** · 2026-05-12 · signed off by Matze after Council Round 3 convergence. Implementation tracked in `agents/roadmaps/road-to-global-first-install.md`.
 
-Originates from user ask: "Es macht keinen Sinn, das paket nicht
+Originates from user ask: "Es macht keinen Sinn, das paket nicht  <!-- md-language-check: ignore -->
 global zu installieren." Validated through AI Council (2 + 1 rounds,
 2026-05-12 · members claude-sonnet-4-5 + gpt-4o). Council session:
 [`agents/runtime/council/sessions/2026-05-12-global-first-strategy/`](../../agents/runtime/council/sessions/2026-05-12-global-first-strategy/). <!-- council-ref-allowed: ADR decision trace -->

package/docs/decisions/ADR-008-installed-tools-manifest.md

@@ -17,7 +17,7 @@ phase: v2.x · post-global-first-install
 
 Originates from user ask (Matze, 2026-05-12): "Sollten wir auf
 Projektebene festhalten, welche Agents wir initialisiert haben, damit
-bei jedem Sync das Verzeichnis aktualisiert werden kann?" Validated
+bei jedem Sync das Verzeichnis aktualisiert werden kann?" Validated  <!-- md-language-check: ignore -->
 through AI Council Round 1 (claude-sonnet-4-5 + gpt-4o, $0.0298 actual,
 both converged on "yes, separate file"). Council session:
 [`agents/runtime/council/sessions/2026-05-12-project-settings-and-v1-v2/`](../../agents/runtime/council/sessions/2026-05-12-project-settings-and-v1-v2/). <!-- council-ref-allowed: ADR decision trace -->

package/docs/decisions/ADR-009-event4u-namespace.md

@@ -13,7 +13,7 @@ phase: v2.4 · namespace-and-claude-desktop
 ## Status
 
 **Accepted** · 2026-05-13 · signed off by Matze after the user ask
-*"auf globaler ebene sollen unsere dateien in einem
+*"auf globaler ebene sollen unsere dateien in einem  <!-- md-language-check: ignore -->
 .event4u/agent-config Folder im user ordner landen"* combined with the
 Claude Desktop deployment gap surfaced in the same turn (the v1
 `marker-only` integration did not actually make any skill visible to
@@ -36,7 +36,7 @@ points surfaced in production use:
    `~/.config/agent-config/claude-desktop/` and called it done. Claude
    Desktop does **not** auto-discover skills from any filesystem path
    — the user must upload each skill through **Settings → Customize →
-   Skills → Upload**. Research against `nextlevelbuilder/ui-ux-pro-max-skill`
+   Skills → Upload**. Research against an external reference suite
    and the Anthropic Skills API docs confirmed there is no public
    bulk-upload API for personal installs (the `/v1/skills` endpoint is
    workspace + code-execution gated).
@@ -180,8 +180,8 @@ zero-data-loss** by design:
 ## References
 
 - [`agents/roadmaps/road-to-event4u-namespace-and-claude-desktop.md`](../../agents/roadmaps/road-to-event4u-namespace-and-claude-desktop.md)
-- [`scripts/_lib/user_global_paths.py`](../../scripts/_lib/user_global_paths.py)
-- [`scripts/_lib/claude_desktop_bundler.py`](../../scripts/_lib/claude_desktop_bundler.py)
+- [`scripts/_lib/user_global_paths.py`](../../src/scripts/_lib/user_global_paths.py)
+- [`scripts/_lib/claude_desktop_bundler.py`](../../src/scripts/_lib/claude_desktop_bundler.py)
 - [`docs/setup/per-ide/claude-desktop.md`](../setup/per-ide/claude-desktop.md)
 - [`docs/migration/v1-to-v2.md`](../migration/v1-to-v2.md) § v2 → v2.4
 - ADR-007 (predecessor — global-first install scopes)