@event4u/agent-config 8.1.0 → 8.2.0

package/dist/agent-src/skills/ai-council/SKILL.md

@@ -123,7 +123,7 @@ THE HOST AGENT IS THE CONVENER, NEVER A REVIEWER.
 
 If you find yourself wanting to "frame" the artefact for the council,
 stop. Framing is exactly what kills the second-opinion value. Use the
-unbiased system prompts in `scripts/ai_council/prompts.py`; do not
+unbiased system prompts in `scripts/ai_council/prompts.ts`; do not
 roll your own.
 
 The host runs the council and synthesises convergence — it is the
@@ -148,7 +148,7 @@ in front of every member's system prompt, assembled by
 | One paragraph of repo purpose from `README.md` (max 400 chars) | Host-agent framing language ("I think this looks weak", "the user probably wants…") |
 | The user's **original ask** verbatim (the free-form sentence that triggered `/council`) | Anything the host agent generated about the artefact |
 
-`detect_project_context()` in `scripts/ai_council/project_context.py`
+`detect_project_context()` in `scripts/ai_council/project_context.ts`
 reads only the manifest files + root README; missing fields collapse
 to `None` and the preamble silently omits the line. With both
 `project=None` and `original_ask=""`, the preamble degrades to the
@@ -167,7 +167,7 @@ travel changes.
 | `manual` | `ManualClient` | no | `stdout` (prompt block) + `stdin` (user pastes the web-UI reply, terminated by a line containing only `END`) | shipped (Phase 2b) |
 | `cli` | `AnthropicCliClient` / `OpenAICliClient` / `GeminiCliClient` | no (subscription-authed) | local subprocess against the vendor CLI (`claude`, `codex`, `gemini`); auth delegated to the CLI's own session, no API key in this process | shipped (anthropic/openai/gemini · Phase 3) |
 
-Resolution lives in `scripts/ai_council/modes.py`:
+Resolution lives in `scripts/ai_council/modes.ts`:
 `resolve_mode(name, invocation_mode, member_settings, global_mode)`
 with precedence **invocation flag > per-member setting > global
 setting > default (`manual`)**. Whitespace-and-case insensitive; empty
@@ -293,15 +293,15 @@ not a default.
 1. **Resolve target.** Identify the artefact mode (`prompt`, `roadmap`,
    `diff`, `files`) and locate the source. Refuse to proceed if the
    target is ambiguous.
-2. **Bundle + redact.** Call `scripts/ai_council/bundler.py` to produce
+2. **Bundle + redact.** Call `scripts/ai_council/bundler.ts` to produce
    a redacted artefact bundle. If `BundleTooLarge` fires, surface the
    size and ask the user to narrow scope — do NOT truncate silently.
 3. **Confirm spend.** Before any network call, surface members + cost
    ceiling and require an explicit user `1` to proceed. Autonomy
    settings do not override this gate.
 4. **Fan out.** Dispatch the bundle to each enabled council member via
-   `scripts/ai_council/orchestrator.py`. Each member receives the
-   neutrality preamble from `prompts.py` plus the artefact — nothing
+   `scripts/ai_council/orchestrator.ts`. Each member receives the
+   neutrality preamble from `prompts.ts` plus the artefact — nothing
    from the host agent's prior reasoning.
 5. **Render results.** Stack each member's response under its own
    provider-attributed heading. Never merge or paraphrase responses
@@ -412,10 +412,10 @@ matching `road-to-<topic-slug>` roadmap under `agents/roadmaps/`).
   that may include council artefacts as part of their evidence trail
   (e.g. `audit-2026-05-14-north-star/` bundles its triggering
   question, raw responses, and synthesis alongside the audit's
-  findings). The layout linter (`scripts/check_council_layout.py`)
+  findings). The layout linter (`scripts/check_council_layout.ts`)
   skips these directories.
 
-`scripts/check_council_layout.py` is the mechanical check for the
+`scripts/check_council_layout.ts` is the mechanical check for the
 output path convention — wire it into the package's CI pipeline so
 violations break the build.
 
@@ -458,7 +458,7 @@ attributed to the host.
 The **Convergence / Divergence** slot in `council:render` output is
 populated with a lens-specific synthesis prompt. The host agent reads
 this prompt and writes the summary in the shape it dictates. The five
-templates live in `scripts/ai_council/prompts.py::_SYNTHESIS_TABLE`
+templates live in `scripts/ai_council/prompts.ts::_SYNTHESIS_TABLE`
 and are exposed via `synthesis_template(mode)`.
 
 **R4 Q4 split** — decision lenses get a structured Karpathy-style
@@ -546,7 +546,7 @@ Real failure modes seen in the wild:
 | Anti-pattern | Why it's wrong | Correct approach |
 |---|---|---|
 | "Pre-warm" the council with the agent's own analysis. | Bias attack — collapses the reviewer to a yes-man. | Send the artefact text only. |
-| Paste the host-agent identity ("I am Augment / Claude Code…") | Identity primes the reviewer's model. | Neutrality preamble in `prompts.py` already handles this. |
+| Paste the host-agent identity ("I am Augment / Claude Code…") | Identity primes the reviewer's model. | Neutrality preamble in `prompts.ts` already handles this. |
 | Silently truncate a too-large bundle. | Misleads the reviewer into thinking they saw the whole thing. | Bundler raises `BundleTooLarge`; surface and ask for narrower scope. |
 | Reuse the same SDK client across calls without re-loading the key. | Leaks the key in long-lived process state. | Each invocation builds fresh clients from `load_*_key()`. |
 | Auto-spend tokens under `personal.autonomy: on`. | Autonomy ≠ permission to spend money. | Always ask before consultation, even under autonomy. |
@@ -586,7 +586,7 @@ per-invocation caps from `ai_council.cost_budget`:
   `daily_budget_exceeded` instead of `cost_budget_exceeded`.
 
 Prices come from `agents/runtime/.agent-prices.md` (gitignored, refreshed weekly).
-The pricing module bootstraps it from `_default_prices.py` on first
+The pricing module bootstraps it from `_default_prices.ts` on first
 use and flags it stale when older than the most recent Monday 00:00
 UTC.
 
@@ -937,16 +937,16 @@ fast-path entries (clean session is not an error).
 - `/council` command — the user-facing entry point.
 - `subagent-orchestration` skill — internal multi-agent variant (no
   network, no spend, but no diversity of weights either).
-- `scripts/ai_council/prompts.py` — neutrality preamble + per-mode
+- `scripts/ai_council/prompts.ts` — neutrality preamble + per-mode
   system prompts.
-- `scripts/ai_council/advisors.py` — replace-mode planning + persona
+- `scripts/ai_council/advisors.ts` — replace-mode planning + persona
   resolution.
-- `scripts/ai_council/bundler.py` — redaction pattern set + size
+- `scripts/ai_council/bundler.ts` — redaction pattern set + size
   guard.
 - `docs/customization.md` § `ai_council.*` — settings reference.
 - `docs/contracts/ai-council-config.md` § advisors — schema + precedence
   contract.
 - `docs/contracts/ai-council-config.md` § Decision-replay artefact —
   Phase 9 audit trail contract + redaction modes.
-- `scripts/ai_council/replay.py` — pure projection renderer (no model
+- `scripts/ai_council/replay.ts` — pure projection renderer (no model
   calls).

package/dist/agent-src/skills/api-design/SKILL.md

@@ -17,7 +17,7 @@ packs:
 
 > **Grounded corpus (Tier-1 consultation):** pagination, versioning,
 > error shape (RFC 9457), idempotency, async ops, rate limiting, bulk,
-> naming, expansion, webhooks — `ground.py search --manifest
+> naming, expansion, webhooks — `./scripts-run <skills-root>/corpus-grounding/scripts/ground search --manifest
 > <skills-root>/api-design/data/manifest.json "<concern>"` proposes the
 > grounded pattern (+ hardening + anti-patterns + RFC link) before
 > designing from memory. Corpus: [`data/api-patterns.csv`](data/api-patterns.csv).

package/dist/agent-src/skills/api-endpoint/SKILL.md

@@ -14,7 +14,8 @@ packs:
 
 > **Grounded corpus:** endpoint-shape decisions (pagination, error shape,
 > idempotency, rate limits) ground via the
-> [`api-design`](../api-design/SKILL.md) corpus — `ground.py search
+> [`api-design`](../api-design/SKILL.md) corpus — `./scripts-run
+> <skills-root>/corpus-grounding/scripts/ground search
 > --manifest <skills-root>/api-design/data/manifest.json "<concern>"`.
 
 ## When to use

package/dist/agent-src/skills/authz-review/SKILL.md

@@ -19,7 +19,7 @@ packs:
 > **Grounded corpus (Tier-1 consultation):** the threat corpus's
 > `authorization` + `tenancy` rows (IDOR, mass-assignment escalation,
 > unscoped queries, sealed job context — each with negative tests) come
-> from `ground.py ground --manifest
+> from `./scripts-run <skills-root>/corpus-grounding/scripts/ground ground --manifest
 > <skills-root>/threat-modeling/data/manifest.json "<check under review>"`.
 > Cite corpus rows in findings; surface the evidence gap when no row.
 

package/dist/agent-src/skills/blade-ui/SKILL.md

@@ -1,7 +1,7 @@
 ---
 model_tier: medium
 name: blade-ui
-description: "Use when the project's frontend stack is Blade — dispatched by `directives/ui/{apply,review,polish}.py`. Covers views, components, partials, layouts, and view logic."
+description: "Use when the project's frontend stack is Blade — dispatched by `directives/ui/{apply,review,polish}.ts`. Covers views, components, partials, layouts, and view logic."
 domain: engineering
 workspaces:
   - engineering
@@ -17,7 +17,8 @@ install:
 # blade-ui
 
 > **Grounded stack guidance:** pull idiomatic Do/Don't + docs URLs from the
-> adopted stack corpus before improvising — `ground.py search --manifest
+> adopted stack corpus before improvising — `./scripts-run
+> <skills-root>/corpus-grounding/scripts/ground search --manifest
 > <skills-root>/design-intelligence/data/manifest.json --stack laravel "<topic>"`
 > (the `laravel` stack file is Blade/UI-scoped, not general Laravel). See
 > [`design-intelligence`](../design-intelligence/SKILL.md).
@@ -25,23 +26,23 @@ install:
 ## Positioning — dispatched, not standalone
 
 `blade-ui` is the **apply-step executor** for the Blade stack. It is
-invoked by [`directives/ui/apply.py`](../../templates/scripts/work_engine/directives/ui/apply.py)
-once the design brief is locked, and revisited by `review.py` /
-`polish.py` during the design-review loop. It does **not** own the
+invoked by [`directives/ui/apply.ts`](../../templates/scripts/work_engine/directives/ui/apply.ts)
+once the design brief is locked, and revisited by `review.ts` /
+`polish.ts` during the design-review loop. It does **not** own the
 flow, does **not** drive the audit, and does **not** lock the design.
 
 | Concern | Owner |
 |---|---|
 | Audit + token inventory (mandatory pre-step) | [`existing-ui-audit`](../existing-ui-audit/SKILL.md) |
-| Design brief (layout / states / microcopy) | [`directives/ui/design.py`](../../templates/scripts/work_engine/directives/ui/design.py) |
+| Design brief (layout / states / microcopy) | [`directives/ui/design.ts`](../../templates/scripts/work_engine/directives/ui/design.ts) |
 | Universal design heuristics | [`fe-design`](../fe-design/SKILL.md) |
-| Review + polish loop | [`directives/ui/review.py`](../../templates/scripts/work_engine/directives/ui/review.py) + [`polish.py`](../../templates/scripts/work_engine/directives/ui/polish.py) |
+| Review + polish loop | [`directives/ui/review.ts`](../../templates/scripts/work_engine/directives/ui/review.ts) + [`polish.ts`](../../templates/scripts/work_engine/directives/ui/polish.ts) |
 
 ## When to use
 
 Cite this skill when:
 
-- `state.stack.frontend == "blade"` (or the project is clearly Blade-only without Livewire / Flux) and `directives/ui/apply.py` dispatches to this skill
+- `state.stack.frontend == "blade"` (or the project is clearly Blade-only without Livewire / Flux) and `directives/ui/apply.ts` dispatches to this skill
 - Editing or creating Blade views, components, partials, layouts, or forms
 
 Do NOT use when:
@@ -92,8 +93,8 @@ Do NOT use when:
 
 ### Review pass — a11y findings + preview envelope
 
-When this skill is dispatched by `directives/ui/review.py` (test slot)
-or `directives/ui/polish.py` (verify slot) — i.e. a review/polish run,
+When this skill is dispatched by `directives/ui/review.ts` (test slot)
+or `directives/ui/polish.ts` (verify slot) — i.e. a review/polish run,
 not the initial apply — it also emits:
 
 - `state.ui_review.a11y` — `{violations: [{rule, selector, severity}, ...],

package/dist/agent-src/skills/blameless-post-mortem/SKILL.md

@@ -124,9 +124,10 @@ An action without a closure criterion is a wish, not a plan.
 
 Per [`docs/contracts/analysis-memory-loop.md § 2`](../../docs/contracts/analysis-memory-loop.md):
 
-1. Run the dedup pre-check — call `find_duplicate(...)` from
-   `scripts.memory_lookup` over the same key-space (incident type,
-   affected paths, decision area).
+1. Run the dedup pre-check — `find_duplicate(...)` from
+   `scripts/memory_lookup.ts` over the same key-space (incident type,
+   affected paths, decision area); a `memory:lookup` query over the
+   same keys is the CLI equivalent.
 2. **Match found** — propose a `frequency` / `supersedes` update to
    the existing entry; do NOT create a new candidate.
 3. **No match** — draft a REDACTED `incident-learnings` candidate to
@@ -143,10 +144,10 @@ Per [`docs/contracts/analysis-memory-loop.md § 2`](../../docs/contracts/analysi
 
 4. NEVER auto-promote. The human drives promotion via `/memory
    promote`. If the candidate fails the admission gate
-   (`check_memory_proposal.py`), surface the gap to the user.
+   (`check_memory_proposal.ts`), surface the gap to the user.
 
-If `retrieve()` returns stale entries in `skipped`, surface them
-explicitly — never silently use stale data.
+If the lookup surfaces stale entries (past `review_after_days`),
+surface them explicitly — never silently use stale data.
 
 ## Output
 
@@ -181,8 +182,8 @@ The post-mortem produces, in order:
   Severity is `near-miss`; the extra two questions apply.
 - **Corrective actions without closure criteria** are non-actionable.
   Always require a specific, testable done condition.
-- **Stale memory entries** surface in `skipped`, not `results`. Never
-  silently reuse stale data.
+- **Stale memory entries** must be surfaced as stale, never merged
+  into the findings. Never silently reuse stale data.
 
 ## See also
 

package/dist/agent-src/skills/code-review/SKILL.md

@@ -105,10 +105,9 @@ The checks below are stack-agnostic. For framework-specific conventions (PSR-12
 
 ## Before creating a PR
 
-1. Run the project's quality pipeline (see the stack carve-out for the exact commands — PHP: `quality-tools`).
-2. Run tests via the project's runner (`make test`, `npm test`, `pytest`, `go test ./...`, or the project's wrapper script).
-3. Ensure CI passes on the branch.
-4. Self-review the diff: `git diff origin/main..HEAD`.
+1. When `quality.local_auto_run: true`: run the project's quality pipeline (see the stack carve-out for the exact commands — PHP: `quality-tools`) and tests via the project's runner (`make test`, `npm test`, `pytest`, `go test ./...`, or the project's wrapper script). Under the default (`false` / missing): skip both — remote CI on the PR is the gate.
+2. Ensure CI passes on the branch (remote CI is authoritative).
+3. Self-review the diff: `git diff origin/main..HEAD`.
 
 ## Receiving feedback
 

package/dist/agent-src/skills/command-routing/SKILL.md

@@ -45,11 +45,11 @@ Only ask the user if inference fails and the command cannot proceed without the
 | `.augment/commands/` | Shared commands (work across projects) |
 | `agents/overrides/commands/` | Project-specific overrides (used instead of original) |
 
-## Commands that dispatch to a Python engine
+## Commands that dispatch to the work engine
 
 Most commands are pure markdown procedures — the agent reads the steps
-and executes them. Two commands delegate to the `work_engine` Python
-module via the `./agent-config` dispatcher; both share the same
+and executes them. Two commands delegate to the `work_engine` engine
+(TypeScript, run via tsx) via the `./agent-config` dispatcher; both share the same
 Option-A loop (read state → run engine → handle exit code → repeat),
 they only differ in the input envelope they build:
 

package/dist/agent-src/skills/command-writing/SKILL.md

@@ -59,7 +59,7 @@ Why optional + not yet blocking, + the upgrade trigger:
 ## Commands ARE Claude skills (projection reality)
 
 Every `src/agent-src/commands/{name}.md` projects to
-`.claude/skills/{slug}/SKILL.md` via `scripts/condense.py`
+`.claude/skills/{slug}/SKILL.md` via `scripts/condense.ts`
 (`generate_claude_commands`). Nested commands flatten with `-`
 (`council/default.md` → `council-default`). Skills + commands share the
 **same `.claude/skills/` namespace** — Claude does not distinguish them.
@@ -205,7 +205,7 @@ multi-paragraph explanation, extract it into a skill and call it.
 ### 6. Governance baseline (when introducing a new linter check)
 
 **Advisory, reviewer-checked — no CI gate.** When the same PR adds a
-new check to `scripts/skill_linter.py` (or strengthens an existing one)
+new check to `scripts/skill_linter.ts` (or strengthens an existing one)
 such that previously-clean commands now warn, the PR body MUST record
 the pre-existing violations on `main` in a Markdown table:
 
@@ -273,7 +273,7 @@ to every command you author.
 ## Cloud Behavior
 
 On cloud surfaces (Claude.ai Web, Skills API) the package's
-`scripts/skill_linter.py`, `scripts/condense.py`, and the `task`
+`scripts/skill_linter.ts`, `scripts/condense.ts`, and the `task`
 runner are not available. This skill still applies — but with
 prose-only validation:
 

package/dist/agent-src/skills/condense-memory/SKILL.md

@@ -29,7 +29,7 @@ Use when:
 
 - Condense reply, commit message, PR body, ticket summary, or any deliverable written *for* human reader — those are carve-outs in [`telegraph-speak § Carve-outs`](../../rules/telegraph-speak.md) and stay verbatim.
 - Condense path matching sensitive-file denylist (`.env*`, `.netrc`, `credentials*`, `secrets*`, `id_rsa*`, `*.pem|key|p12|pfx|crt|cer|jks`, `.ssh/*`) — script refuses with `SensitivePathError` and so should you.
-- Condense generated file (`dist/agent-src/`, `.augment/`, `.claude/`, `.cursor/`, `.clinerules/`, `.windsurfrules`) — edit source in `.agent-src.uncondensed/` and regenerate via package's sync + generate-tools scripts (`scripts/condense.sh --sync` + `scripts/condense.py --generate-tools`).
+- Condense generated file (`dist/agent-src/`, `.augment/`, `.claude/`, `.cursor/`, `.clinerules/`, `.windsurfrules`) — edit source in `.agent-src.uncondensed/` and regenerate via package's sync + generate-tools scripts (`scripts/condense.sh --sync` + `scripts/condense.ts --generate-tools`).
 - Hand-edit condensed memory file in place — run `--decondense` first; next condense pass refuses on body-hash drift (`CondensationRefused`).
 - Commit condensed file without committing matching `.original.md` backup — round-trip breaks otherwise.
 
@@ -99,7 +99,7 @@ CLI exit codes:
 - **Denylist false positive** — sensitive-looking filename outside denylist surface (project-specific naming) will still pass `assert_safe()`. Denylist necessary but not sufficient; maintainer responsible for never feeding secrets to condenseor.
 - **Frontmatter ordering with existing keys** — if target already has frontmatter, condenseor preserves existing keys, drops any prior `original_sha256:` / `condensed_at:` entries, and appends new pair. Other agents reading file should treat SHA + timestamp pair as canonical condensation marker, not file size.
 - **Negative savings on pointer-heavy files** — `templates/AGENTS.md` already following Thin-Root (≥ 40 % pointers, ≥ 60-char *why*-clauses) has little prose left to drop; condensation may net near-zero or even add bytes via frontmatter. Run [`agents-md-thin-root`](../agents-md-thin-root/SKILL.md) first to maximise pointer share, then measure whether this skill still pays.
-- **Generated-tree drift** — condensing `.agent-src.uncondensed/templates/AGENTS.md` does NOT propagate to `.augment/`, `.claude/`, etc. until package's sync + generate-tools scripts run (`scripts/condense.sh --sync` + `scripts/condense.py --generate-tools`). Always regenerate after condensing templated file.
+- **Generated-tree drift** — condensing `.agent-src.uncondensed/templates/AGENTS.md` does NOT propagate to `.augment/`, `.claude/`, etc. until package's sync + generate-tools scripts run (`scripts/condense.sh --sync` + `scripts/condense.ts --generate-tools`). Always regenerate after condensing templated file.
 
 ## Measurement — when to condense
 

package/dist/agent-src/skills/context-authoring/SKILL.md

@@ -156,7 +156,7 @@ If the context file declares its own `load_context:` (chain reasoning),
 use logical names rooted at the source — `contexts/<area>/<file>.md`
 for package material or `agents/settings/contexts/<file>.md` for project-local
 material. The `.agent-src.uncondensed/` prefix is rejected by the
-schema regex and by `scripts/lint_load_context.py`. Body links to
+schema regex and by `scripts/lint_load_context.ts`. Body links to
 `docs/guidelines/...` use the verbatim `../../docs/...` relative form.
 Canonical reference: `rule-writing` § 3b and
 `docs/contracts/load-context-schema.md`.

package/dist/agent-src/skills/context-document/SKILL.md

@@ -108,7 +108,7 @@ human curates), a card is governed: it is a cache of *expensive* remote evidence
 **negative facts + pointers** (`trust: durable`); its positive structure is a
 per-line, last-verified **hypothesis** that loads as "Assumed (from card)" and
 must be re-confirmed against the live source before use. Cards pass the
-`check_knowledge_cards.py` pointer-CI (size ≤ 150, mandatory authoritative
+`check_knowledge_cards.ts` pointer-CI (size ≤ 150, mandatory authoritative
 pointer, trust tagging, multi-evidence consistency). See
 [`source-discovery`](../source-discovery/SKILL.md) for when a structure is
 card-worthy and [`evidence-discipline`](../../agent-src/contexts/execution/evidence-discipline.md)

package/dist/agent-src/skills/copilot-agents-optimization/SKILL.md

@@ -156,7 +156,7 @@ about the project** over time:
 Before deduplicating, run three scans:
 
 1. **Legacy identifier scan** — compare both files against the package's
-   `FORBIDDEN_IDENTIFIERS` blocklist (see `scripts/check_portability.py`)
+   `FORBIDDEN_IDENTIFIERS` blocklist (see `scripts/check_portability.ts`)
    plus any project names from `agents/` module docs that don't match
    the current project.
 2. **Stack coherence scan** — auto-detect the actual stack from

package/dist/agent-src/skills/corpus-grounding/SKILL.md

@@ -36,7 +36,8 @@ frontend-hardcoded axis moved into the manifest. Full license obligations:
 - A skill needs **pre-action selection** grounded in curated knowledge
   ("which layout pattern / threat class / index strategy applies here").
 - You are authoring a **new domain corpus** — write a manifest + CSVs,
-  validate with `ground.py validate`, never fork the engine.
+  validate with the grounding CLI (`ground validate` via ./scripts-run),
+  never fork the engine.
 - You need stack-scoped Do/Don't guidance (`--stack react …`).
 
 ## Do NOT
@@ -92,7 +93,8 @@ Stable v1 surface:
    search_stack, evaluate_rules, ground, persist_grounding}`,
    `schema_validator.{load_manifest, validate_manifest, resolve_data_path,
    ManifestError, MANIFEST_VERSION, TIERS}`.
-2. **CLI** — `ground.py {search|ground|validate}` with the flags shown above.
+2. **CLI** — the grounding CLI (`ground {search|ground|validate}` via
+   ./scripts-run) with the flags shown above.
 3. **Result shape** — search: `{domain|stack, query, file, count, results,
    scores, filtered_from, confidence{label,score}, evidence_gap[]}`;
    ground: `{domain, query, category, rule, rules_evaluation{matched,
@@ -138,7 +140,7 @@ Each domain ships `data/manifest.json` beside its CSVs — declaring its
     "priority_domain": "style",
     "name_columns": { "style": "Style Category" },
     "plan": { "style": 3, "color": 2 },
-    "rules_module": "rules.py"
+    "rules_module": "rules.ts"
   },
   "owner": "package-maintainer",
   "refresh_cadence": "quarterly",
@@ -151,8 +153,10 @@ Each domain ships `data/manifest.json` beside its CSVs — declaring its
 - `owner`, `refresh_cadence`, `upstream{repo,sha,last_checked}` are
   **required** — provenance discipline per ADR-061 §6; an unowned corpus
   is not merged.
-- `rules_module` (optional Python escape hatch where JSON rules cap out)
-  must live beside the manifest — absolute paths and `..` are refused.
+- `rules_module` (optional TypeScript escape hatch where JSON rules cap
+  out — a module exposing `evaluate(rules, query, context)`; a legacy
+  `.py` name in a manifest is remapped to its `.ts`/`.js` twin) must live
+  beside the manifest — absolute paths and `..` are refused.
 - Decision rules are surfaced as `matched` **and** `unmatched` — the full
   rule space stays auditable, never a hidden gate.
 
@@ -162,7 +166,7 @@ Each domain ships `data/manifest.json` beside its CSVs — declaring its
    row count; grounding must happen **before** action; "fits in 5 lines →
    it's a rule, not a corpus".
 2. Write `data/manifest.json` + CSVs in your domain skill; run
-   `ground.py validate`.
+   the grounding CLI (`ground validate` via ./scripts-run).
 3. Name an owner + refresh cadence in the manifest, pin the upstream SHA.
 4. Cite this skill from your domain skill; consult via the CLI (Tier-1
    consultation default — propose grounded options, human confirms).
@@ -189,9 +193,9 @@ Each domain ships `data/manifest.json` beside its CSVs — declaring its
   under the caller-chosen `DIR`; nothing else writes.
 - **No network, no subprocess, no secrets** — pure stdlib; embeddings /
   remote retrievers are intentionally not implemented (ADR-061 §2).
-- `rules_module` executes manifest-adjacent Python — same trust domain as
-  the skill that ships the manifest; containment enforced by
-  `resolve_data_path`.
+- `rules_module` executes a manifest-adjacent TypeScript module (dynamic
+  `import()`) — same trust domain as the skill that ships the manifest;
+  containment enforced by `resolve_data_path`.
 
 ## See also
 
@@ -201,5 +205,6 @@ Each domain ships `data/manifest.json` beside its CSVs — declaring its
   architecture, qualification rubric, fork resolutions.
 - [`docs/contracts/skill-bundled-assets.md`](../../../docs/contracts/skill-bundled-assets.md)
   — how bundled `scripts/` + `data/` reach consumer runtime.
-- Tests: `tests/test_corpus_grounding_engine.py`,
-  `tests/test_skill_bundled_assets.py`.
+- Tests: `tests/scripts/skills_corpus_grounding_ground.test.ts`,
+  `tests/scripts/skills_corpus_grounding_bm25_search.test.ts`,
+  `tests/scripts/skills_corpus_grounding_schema_validator.test.ts`.

package/dist/agent-src/skills/corpus-grounding/scripts/ground.ts

@@ -22,7 +22,7 @@
 
 import * as fs from 'node:fs';
 import * as path from 'node:path';
-import { pathToFileURL } from 'node:url';
+import { fileURLToPath, pathToFileURL } from 'node:url';
 
 import {
     PyFloat,
@@ -524,9 +524,30 @@ function _jsonStr(s: string): string {
     return `${out}"`;
 }
 
+function _isCliEntry(): boolean {
+    if (process.argv[1] === undefined) {
+        return false;
+    }
+    const argvUrl = pathToFileURL(path.resolve(process.argv[1])).href;
+    if (import.meta.url === argvUrl) {
+        return true;
+    }
+    // A symlinked invocation (e.g. via an installed `.augment/` projection,
+    // or macOS /var → /private/var temp dirs) makes the raw URLs differ:
+    // import.meta.url is the resolved real path while argv[1] keeps the
+    // symlink path. Compare realpaths so the entry guard still fires
+    // (without this the CLI silently no-ops when run through a symlink).
+    try {
+        const here = fs.realpathSync(fileURLToPath(import.meta.url));
+        const argv = fs.realpathSync(path.resolve(process.argv[1]));
+        return here === argv;
+    } catch {
+        return false;
+    }
+}
+
 const _isMain =
-    process.argv[1] !== undefined &&
-    import.meta.url === pathToFileURL(path.resolve(process.argv[1])).href;
+    _isCliEntry();
 if (_isMain) {
     main()
         .then((code) => {

package/dist/agent-src/skills/dashboard-design/SKILL.md

@@ -27,7 +27,8 @@ Do NOT use when:
 2. **Pick the surface** — Use the decision tables below to choose Grafana, app dashboard, or embed; document audience and refresh cadence.
 3. **Draft the layout** — Sketch panels, choose visualization per signal (RED / USE / Golden Signals), define filters and thresholds.
    **Ground the chart-type choice** in the adopted data-viz corpus instead
-   of memory: `ground.py search --manifest
+   of memory: `./scripts-run <skills-root>/corpus-grounding/scripts/ground
+   search --manifest
    <skills-root>/design-intelligence/data/manifest.json --domain chart
    "<data shape>"` returns the best chart type, when-NOT-to-use, data-volume
    threshold, a11y grade + colorblind fallback, and a library

package/dist/agent-src/skills/database/SKILL.md

@@ -15,7 +15,8 @@ packs:
 
 
 > **Grounded corpus (Tier-1 consultation):** symptom → index/strategy
-> decisions come grounded — `ground.py search --manifest
+> decisions come grounded — `./scripts-run
+> <skills-root>/corpus-grounding/scripts/ground search --manifest
 > <skills-root>/database/data/manifest.json "<symptom>"` returns root
 > cause, strategy, good-code sketch, anti-pattern, and the verification
 > probe (EXPLAIN expectation). Corpus:

package/dist/agent-src/skills/description-assist/SKILL.md

@@ -19,8 +19,8 @@ packs:
   "will this ever fire", "rewrite the description".
 * Called from within the Draft phase of one of the writing skills
   (`skill-writing`, `rule-writing`, `command-writing`, `guideline-writing`).
-* Auditing a skill flagged by `audit_skill_descriptions.py` as `too-short`,
-  `no-trigger-prefix`, or `description_too_long` in `skill_linter.py`.
+* Auditing a skill flagged by `audit_skill_descriptions.ts` as `too-short`,
+  `no-trigger-prefix`, or `description_too_long` in `skill_linter.ts`.
 
 Do NOT use this skill when:
 
@@ -65,7 +65,7 @@ If the target is a skill, check for a trigger-eval report:
 
 * `.agent-src.uncondensed/skills/{name}/evals/triggers.json` — the expected-trigger corpus
 * `evals/last-run.json` — the most recent runner output (see
-  [`scripts/skill_trigger_eval.py`](../../../scripts/skill_trigger_eval.py))
+  [`scripts/skill_trigger_eval.ts`](../../../scripts/skill_trigger_eval.ts))
 
 If `last-run.json` exists and contains failed queries for this skill,
 extract the failure patterns and **use them to inform step 3 variants**.
@@ -159,14 +159,14 @@ and stop. Do not loop further.
 * Do NOT propose more than 3 variants in one turn
 * Do NOT change the skill name, body, or trigger phrases in the same turn
 * Do NOT silently strip or reword quotes inside the description value
-* Do NOT run `scripts/skill_trigger_eval.py` from inside this skill — eval
+* Do NOT run `scripts/skill_trigger_eval.ts` from inside this skill — eval
   execution spends API tokens and is a separate user action
 
 ## Cloud Behavior
 
 On cloud surfaces (Claude.ai Web, Skills API) the package's
-`scripts/audit_skill_descriptions.py`, `scripts/skill_linter.py`,
-and `scripts/skill_trigger_eval.py` are not reachable. The skill
+`scripts/audit_skill_descriptions.ts`, `scripts/skill_linter.ts`,
+and `scripts/skill_trigger_eval.ts` are not reachable. The skill
 still applies — with prose-only inspection:
 
 * Reason from the description text in the conversation. The agent

package/dist/agent-src/skills/design-intelligence/SKILL.md

@@ -26,9 +26,9 @@ execution:
 > consult this corpus FIRST and pre-fill the brief candidates — then the
 > human confirms (`design_confirmed`). Corpus output is a **constraint
 > set**, never final microcopy; the placeholder lock in
-> `directives/ui/design.py` is unaffected.
+> `directives/ui/design.ts` is unaffected.
 >
-> **Boundary (council-locked):** `design.py` stays a pure orchestration
+> **Boundary (council-locked):** `design.ts` stays a pure orchestration
 > gate and never imports the engine — the corpus call lives HERE, in the
 > skill layer, keeping the engine an optional dependency.
 
@@ -221,7 +221,8 @@ consistency; on a fresh session, re-hydrate `state.ui_design` from
 The `review`/`polish` steps gate on `state.ui_review.a11y`. Ground two
 finding classes instead of ad-hoc judgment:
 
-- **Chart-type findings** — `…/ground.py search --manifest … --domain
+- **Chart-type findings** — the grounding CLI (`ground` via
+  ./scripts-run) — `…/ground search --manifest … --domain
   chart "<data shape>"` → `Accessibility Grade`, `A11y Fallback`,
   `Color Guidance` columns justify "wrong chart type / missing colorblind
   fallback" findings with a citable row.
@@ -260,7 +261,7 @@ guidance + docs URLs from here instead of memory.
 
 - Do NOT let the corpus write microcopy — it supplies constraint sets;
   final strings are agent-written (placeholder lock stays in force).
-- Do NOT import the engine into `directives/ui/design.py` — council
+- Do NOT import the engine into `directives/ui/design.ts` — council
   boundary; the corpus call lives in this skill layer.
 - Do NOT propose a new component the `existing-ui-audit` inventory
   already covers — audit findings outrank corpus suggestions.
@@ -273,7 +274,7 @@ guidance + docs URLs from here instead of memory.
 | What already exists (components, tokens) | [`existing-ui-audit`](../existing-ui-audit/SKILL.md) — mandatory pre-step; audit findings outrank corpus suggestions |
 | What to build (grounded selection) | **this skill** |
 | Stack-agnostic heuristics + flow | [`fe-design`](../fe-design/SKILL.md) — invokes this skill for grounding |
-| Orchestration gates + locks | `directives/ui/{design,review,polish}.py` — never import the engine |
+| Orchestration gates + locks | `directives/ui/{design,review,polish}.ts` — never import the engine |
 | WCAG audit method | [`accessibility-auditor`](../accessibility-auditor/SKILL.md) |
 | Token authoring | [`design-tokens`](../design-tokens/SKILL.md) |
 | Lo-fi structure exploration (pre-selection) | [`wireframe`](../wireframe/SKILL.md) — disposable greyscale variants |

package/dist/agent-src/skills/design-tokens/SKILL.md

@@ -21,14 +21,14 @@ execution:
 
 > Token *authoring* skill: the 3-layer DTCG model
 > (**primitive → semantic → component**) with light/dark theming, a
-> Python toolchain (`scripts/tokens.py` — generate / validate / embed,
-> no Node dependency per ADR-061), and a starter template. Selection of
+> TypeScript toolchain (`scripts/tokens.ts` — generate / validate /
+> embed, run via ./scripts-run), and a starter template. Selection of
 > *which* token values fit the product comes grounded from
 > [`design-intelligence`](../design-intelligence/SKILL.md) (WCAG-adjusted
 > color sets, typography pairings); this skill turns the selection into a
 > maintained token system.
 
-Toolchain provenance: Python port of the upstream `.cjs` trio
+Toolchain provenance: port of the upstream `.cjs` trio
 (`generate-tokens`, `validate-tokens`, `embed-tokens`) from
 `nextlevelbuilder/ui-ux-pro-max-skill` `design-system` sub-skill
 @ `b7e3af80f6e331f6fb456667b82b12cade7c9d35` (MIT, last checked
@@ -59,7 +59,7 @@ under `dark.semantic.*` overriding the same semantic names (emitted as a
 (DTCG). Start from
 [the bundled starter template](templates/design-tokens-starter.json).
 
-## Toolchain (`scripts/tokens.py` — skill-relative, any cwd)
+## Toolchain (`scripts/tokens.ts` — skill-relative, any cwd)
 
 ```bash
 # tokens.json → CSS variables (primitives + semantic + components + .dark)
@@ -140,4 +140,4 @@ four-operation split assigns to *rules/linters*, not the corpus).
 - [`tailwind-engineer`](../tailwind-engineer/SKILL.md) — utility-discipline consumer.
 - [`react-shadcn-ui`](../react-shadcn-ui/SKILL.md) — shadcn token conventions.
 - [`docs/guidelines/design-antipatterns.md`](../../../docs/guidelines/design-antipatterns.md) — when authoring the colour layer, avoid the C5 cream/sand default palette (OKLCH L 0.84–0.97, C < 0.06) and C1 purple/violet primaries unless the brand explicitly defines them; the `brand-consistency` rule validates emitted tokens against the active brand profile.
-- Tests: `tests/test_design_tokens_toolchain.py`.
+- Tests: `tests/scripts/skills_design_tokens_tokens.test.ts`.