geneseed 3.3.0 → 3.4.0

package/src/skills/teach.md

@@ -0,0 +1,50 @@
+# {{SKILL}}: teach
+
+> {{DESC_TEACH}}
+
+<!-- invocation: user -->
+
+**Trigger:** the user wants to *learn* something and asks you to teach — "teach me X",
+"get me up to speed on X", "this is confusing, break it down", "explain this passage",
+"how do I get to X in N days". Three shapes, picked from the ask: a **crash course**
+(from zero to functional, fast), a **decode** (confusing material handed over, make it
+click), or a **path** (a specific result by a deadline, day by day). To *test* what the
+user already knows rather than teach it, use the [quiz {{SKILL}}](quiz.md); to gather
+facts from the web first, [research](research.md); to convert a PDF or web page into
+readable form, [ingest](ingest.md).
+
+## Procedure
+1. Pin the footing before teaching anything: what the user wants to be able to *do*
+   afterwards, how much time they have, and what they already know. Ask once if any of
+   the three is missing, then proceed — do not plan around "learn X in general".
+2. Pick the shape and run it:
+   - **Crash course** — act as a teacher who has only this one session and will never
+     see them again; the sole objective is *functional* before time runs out. Name
+     three things explicitly: **what to learn first**, **what to ignore completely**
+     (for now), and **the single exercise that, done once, puts them ahead of most
+     people who have studied this for months**. No theory without immediate use, no
+     bare lists. Have them do the exercise — or set it up so they can — rather than
+     read about it.
+   - **Decode** — read what they are stuck on and, before explaining anything, find the
+     one keystone sentence that makes the rest fall into place. Explain only that
+     sentence first, with an everyday analogy and no technical terms; confirm it
+     landed; then let the rest follow from it instead of re-explaining every line.
+   - **Path** — build a day-by-day route to the stated result. Each day carries exactly
+     three things: one task that fits the stated time budget (default 45 minutes), a
+     concrete criterion so they can tell they did it right, and one explicit
+     *what-not-to-do* that day. Trace the whole path against the goal; if finishing
+     every day would not deliver the result, rebuild it until it does.
+3. Check it took: ask three questions only someone who understood could answer (a
+   crash course or decode), or walk the first day together (a path). A miss goes back
+   to the analogy or the exercise — re-anchor, don't hand over the answer.
+4. Close by naming what to practise next and the gap most likely to trip them, so the
+   learning continues without you; offer the [quiz {{SKILL}}](quiz.md) for when they
+   want it tested.
+
+## Done when
+- The user is functional (crash course), has the keystone idea in plain terms and
+  passed the three questions (decode), or holds a dated path whose every day has a
+  timed task, a criterion and a what-not-to-do (path) — and knows what to practise
+  next.
+
+<!-- INCLUDE: skills/_self-improvement.md -->

package/src/skills/tickets.md

@@ -30,15 +30,4 @@ tracker issues that fresh sessions can pick up independently.
 - Every slice of the spec is a published ticket with explicit blockers, each
   sized for one session, and at least one unblocked ticket exists to start on.
 
-## Self-improvement
-
-Close each run with one beat of reflection on the {{SKILL}} itself:
-- A step misled, a needed step was missing, or the trigger fired wrongly — that
-  is a flaw in this file. Propose the exact edit (trigger, procedure, or
-  done-when) and apply it with the user's assent ({{LAW}} II).
-- A lesson that is *not* a flaw in this file goes to {{MEMORY}} only if it
-  clears {{DOCTRINE}} process 1's bar: it would change how a future session behaves, and a
-  fresh read of the repo would not re-derive it. Update an existing memory over
-  adding one; when in doubt, leave it out.
-- No friction, nothing learned — move on; this loop earns no ceremony. Most
-  runs end here.
+<!-- INCLUDE: skills/_self-improvement.md -->

package/src/skills/wayfinder.md

@@ -35,15 +35,4 @@ the destination can be named but the route to it cannot; the user says
 - Every decision ticket is closed, the map indexes them all, and someone could
   pursue the destination without a single further decision.
 
-## Self-improvement
-
-Close each run with one beat of reflection on the {{SKILL}} itself:
-- A step misled, a needed step was missing, or the trigger fired wrongly — that
-  is a flaw in this file. Propose the exact edit (trigger, procedure, or
-  done-when) and apply it with the user's assent ({{LAW}} II).
-- A lesson that is *not* a flaw in this file goes to {{MEMORY}} only if it
-  clears {{DOCTRINE}} process 1's bar: it would change how a future session behaves, and a
-  fresh read of the repo would not re-derive it. Update an existing memory over
-  adding one; when in doubt, leave it out.
-- No friction, nothing learned — move on; this loop earns no ceremony. Most
-  runs end here.
+<!-- INCLUDE: skills/_self-improvement.md -->

package/src/skills/wiki.md

@@ -54,15 +54,4 @@ truth.
 - anything written follows the wiki's `conventions`, links into the graph,
   sits in the right folder (or the `inbox`), and touches nothing `protected`.
 
-## Self-improvement
-
-Close each run with one beat of reflection on the {{SKILL}} itself:
-- A step misled, a needed step was missing, or the trigger fired wrongly — that
-  is a flaw in this file. Propose the exact edit (trigger, procedure, or
-  done-when) and apply it with the user's assent ({{LAW}} II).
-- A lesson that is *not* a flaw in this file goes to {{MEMORY}} only if it
-  clears {{DOCTRINE}} process 1's bar: it would change how a future session behaves, and a
-  fresh read of the repo would not re-derive it. Update an existing memory over
-  adding one; when in doubt, leave it out.
-- No friction, nothing learned — move on; this loop earns no ceremony. Most
-  runs end here.
+<!-- INCLUDE: skills/_self-improvement.md -->

package/src/skills/workflow.md

@@ -2,37 +2,85 @@
 
 > {{DESC_WORKFLOW}}
 
-**Trigger:** a task that benefits from *deterministic* multi-agent orchestration — fan-out across independent units, a staged find→verify pipeline, or a loop that accumulates to a target — **and** the host provides the `workflow` tool (OpenCode). When the host has no such tool, use [parallel-agents](parallel-agents.md) or [council](council.md) instead — those are *model-driven*; this {{SKILL}} is *code-driven*.
+**Trigger:** a task that benefits from *deterministic* multi-agent orchestration — a
+fan-out across independent units, a staged find→verify pipeline, a fresh-context
+handoff between phases, or a loop that accumulates to a target — **and** the host
+exposes a workflow tool. When it does not, use [parallel-agents](parallel-agents.md)
+or [council](council.md) instead — those are *model-driven*; this {{SKILL}} is
+*code-driven*.
 
 ## What it is
 
-A `workflow` is a saved script that orchestrates subagents in **code**, not prose. The script — not the model — decides what fans out, what runs in sequence, and what verifies. The host runs it and hands you back the distilled result. Use it when the control flow should be exact and repeatable rather than re-improvised each time.
+A workflow is a script that orchestrates subagents in **code**, not prose. The
+script — not the model — decides what fans out, what runs in sequence, and what
+verifies; the host runs it and hands back a distilled result. Use it when the control
+flow should be exact and repeatable rather than re-improvised each time, and when the
+work is worth a crew: a one-file tweak or a question never is.
 
-Saved workflows live in the host's workflows directory — on OpenCode that is `.opencode/workflows/` in the repo, or `workflows/` under the global config (see the adapter README) — and you run one **by name**, you do not author one inline. To add a new workflow, copy an existing script and register it there.
+## The hosts
+
+| Host | Tool | Scripts | Opt-in |
+|---|---|---|---|
+| OpenCode | `workflow` (the `geneseed-workflow` plugin) | **saved only** — `<name>.js` in `.opencode/workflows/` per repo or `<config>/workflows/` global (`GENESEED_WORKFLOWS_DIR` overrides); nothing inline is eval'd | none — call it when the shape fits |
+| Claude Code | `Workflow` (native) | a saved `.claude/workflows/<name>` by `name`, **or** an inline `script` you author | **required** — the tool refuses unless the user opted in for this task ("use a workflow", "ultracode", a skill that says so) |
+| Bob, Copilot | none | — | fall back to [parallel-agents](parallel-agents.md) / [council](council.md) |
+
+Both runtimes speak the same primitives — `agent()`, `parallel()`, `pipeline()`,
+`phase()`, `args` — and every script opens with a literal `export const meta = { name,
+description, phases }`. The four shipped scripts are OpenCode-saved today; on Claude
+Code the same shapes are authored inline (the host's `workflow-authoring` reference
+carries the script API — load it before writing one).
+
+## The shipped shapes
+
+- **review** — sweep a change across dimensions, then adversarially verify each
+  finding before reporting (the canonical find→verify pipeline: a dimension's findings
+  verify as soon as that dimension returns, no barrier).
+- **research-plan-implement** — three clean phases, each in its own child context,
+  carrying only the distilled output of the prior phase.
+- **council** — the [council](council.md) debate on rails: seat the stance {{AGENTS}},
+  gather positions in parallel, chair synthesises, dissent preserved.
+- **dispatch** — decompose a multi-domain goal, route each subtask to its owning
+  capability {{AGENT}}, converge (model-driven fallback: [parallel-agents](parallel-agents.md)).
 
 ## Procedure
-1. Confirm the host exposes the `workflow` tool. If not, fall back to [parallel-agents](parallel-agents.md) / [council](council.md) and stop here.
-2. Pick the saved workflow that fits the shape of the work — call `workflow` with no `name`, or an unknown one, to list what is available:
-   - **review** — sweep a change across dimensions, then adversarially verify each finding before reporting (the canonical find→verify pipeline).
-   - **research-plan-implement** — three clean phases with fresh-context handoffs between them.
-   - **council** — the [council](council.md) debate as deterministic code: seat the stance {{AGENTS}}, gather positions in parallel, synthesise a verdict.
-   - **dispatch** — decompose a multi-domain goal, route each subtask to its owning capability {{AGENT}}, converge the results (the live-orchestrator shape; model-driven fallback: [parallel-agents](parallel-agents.md)).
-3. Run it: `workflow({ name, args })`. Pass the task-specific inputs (target paths, the motion, the question) as `args` — the script reads them.
-4. Read the returned summary. The full structured result and a phase-by-phase trace are written to the run's progress file; point the user at it if they want the detail.
-5. Act on the result yourself — the workflow gathers and verifies, but committing, pushing, or merging stays with you ({{DOCTRINE}} process 5).
+1. **Confirm the host and the opt-in.** OpenCode: the `workflow` tool is present. Claude
+   Code: the tool is present *and* the user opted in — if the shape plainly warrants a
+   workflow but they have not, describe it in two lines with its rough cost (how many
+   agents) and ask; never call the tool on your own judgement. No tool at all → fall
+   back and stop here.
+2. **Confirm the task is crew-sized** and the shape is one of the four above (or a
+   plain variation). Sequential work with shared state is a [plan](plan.md); a
+   dev↔test loop in an isolated tree is a [pipeline](pipeline.md), which itself runs on
+   a workflow where one exists.
+3. **Pick saved over inline.** List what is available first — OpenCode: call `workflow`
+   with no `name`; Claude Code: read `.claude/workflows/`. A saved script has been run
+   before; an inline one has not. Author inline only when no saved shape fits, keep it
+   to the primitives, and hold to the host's size guideline unless the user asked for
+   more.
+4. **Pass the task as `args`, never baked in** — target paths, the motion, the question,
+   any timestamp. Scripts stay deterministic: no wall clock, no randomness, nothing
+   read from the environment that the next run would see differently, so a run is a
+   pure function of (script, args, child replies) and can be replayed.
+5. **Run it and wait for the result, not the transcript.** The run is asynchronous on
+   Claude Code (a task notification arrives; `/workflows` watches it live) and
+   synchronous on OpenCode. Read the returned summary; the full structured result and
+   the phase-by-phase trace are in the run's progress file — point the user there for
+   detail, do not paste it. Never fabricate a result for a run that has not returned.
+6. **On failure, resume — don't rerun.** A script that died mid-way keeps its completed
+   `agent()` results; fix the script or the args and resume from the run id where the
+   host supports it, so finished stages are not paid twice. A stage that keeps failing
+   is a finding about the task, not a reason to loop.
+7. **Exit — carry the conclusion yourself.** The workflow gathers and verifies;
+   committing, pushing, merging, or opening anything outward stays with you
+   ({{DOCTRINE}} process 5, {{LAW}} IV). State what the run concluded, what you are
+   doing with it, and where the trace lives.
 
 ## Done when
-- The right saved workflow ran to completion, its result was read, and you have carried its conclusion forward — or, where no `workflow` tool exists, the equivalent model-driven {{SKILL}} was used instead.
-
-## Self-improvement
-
-Close each run with one beat of reflection on the {{SKILL}} itself:
-- A step misled, a needed step was missing, or the trigger fired wrongly — that
-  is a flaw in this file. Propose the exact edit (trigger, procedure, or
-  done-when) and apply it with the user's assent ({{LAW}} II).
-- A lesson that is *not* a flaw in this file goes to {{MEMORY}} only if it
-  clears {{DOCTRINE}} process 1's bar: it would change how a future session behaves, and a
-  fresh read of the repo would not re-derive it. Update an existing memory over
-  adding one; when in doubt, leave it out.
-- No friction, nothing learned — move on; this loop earns no ceremony. Most
-  runs end here.
+- The right shape ran to completion on the host's tool (with the user's opt-in where
+  the host demands one), its summary was read and its trace is locatable, a failed run
+  was resumed rather than restarted, and you — not the script — have carried its
+  conclusion into the next {{SKILL}} or the commit gate. Or: no tool existed, and the
+  model-driven equivalent ran instead.
+
+<!-- INCLUDE: skills/_self-improvement.md -->

package/src/skills/worktree.md

@@ -0,0 +1,52 @@
+# {{SKILL}}: worktree
+
+> {{DESC_WORKTREE}}
+
+**Trigger:** work must not share a working tree with what is already in flight — a
+[pipeline](pipeline.md) crew needs its own tree, a second task starts while the first
+is uncommitted, another agent or session is editing this checkout, a risky migration
+or rewrite needs a disposable copy, or the user says "worktree", "isolate this",
+"don't touch my checkout". Also the first thing to reach for when `git status` shows
+changes you did not make.
+
+## Procedure
+1. **Look before you branch** ({{LAW}} III). `git status` and `git worktree list`
+   first. Uncommitted changes you did not author mean someone else owns this tree —
+   never `checkout -b` from it (that moves *their* HEAD onto your branch) and never
+   stash or revert it. Note the branch they are on; you will not touch it.
+2. **Create the tree beside the repo, from a clean base:**
+   ```
+   git worktree add -b <type>/<slug> ../<repo>-<slug> <base>
+   ```
+   `<base>` is the integration branch (`main`, or `origin/main` after a `git fetch`),
+   never the dirty tree's HEAD. One worktree per task; name the directory after the
+   branch so `git worktree list` reads as a task list.
+3. **Make it runnable.** A worktree shares the repository, not the untracked files:
+   install dependencies (`npm ci`, the project's equivalent) or link the ones that are
+   large and pure; copy no `.env` — secrets are the user's to place ({{LAW}} I). Run
+   the project's checks once in the new tree and read the output before changing
+   anything, so a later red is yours.
+4. **Work only in the worktree.** Every edit, test, commit and `doctor`/lint run
+   happens under its path — use absolute paths in tools whose cwd resets. The
+   original checkout stays as found. Host hooks or shims that write machine-wide
+   state (a global config, a hook shim) run from the worktree only when they resolve
+   the repo through it; verify with the tool's own status command rather than assume.
+5. **Integrate through the normal gate** — commit via the [commit {{SKILL}}](commit.md),
+   ship via [ship](ship.md). A worktree earns no exemption from per-commit and
+   per-push consent ({{DOCTRINE}} process 5). Expect conflicts with a sibling tree that
+   touched the same files; resolve them on the branch, not in the other tree.
+6. **Clean up when the branch is merged or abandoned** — with consent, since removal
+   is deletion ({{LAW}} IV):
+   ```
+   git worktree remove ../<repo>-<slug>
+   git branch -d <type>/<slug>
+   ```
+   `git worktree prune` after a directory was deleted by hand. Leave nothing listed
+   that no task owns.
+
+## Done when
+- The task ran to its commit or PR entirely inside its own tree from a clean base,
+  the original checkout is byte-for-byte as it was found, and the worktree is either
+  still listed with a live task or removed with its branch.
+
+<!-- INCLUDE: skills/_self-improvement.md -->

package/themes/_TEMPLATE.json

@@ -23,6 +23,7 @@
   "DESC_DOCS": "<one-line description: writes user-facing docs after code lands>",
   "DESC_SECURITY": "<one-line description: audits changes touching the security surface>",
   "DESC_EXPLORER": "<one-line description: reads/searches widely, returns only distilled findings>",
+  "DESC_RESEARCHER": "<one-line description: searches and fetches the open web, cross-checks every claim, returns sourced findings>",
   "DESC_ADVOCATE": "<council seat: argues the strongest case FOR the proposal>",
   "DESC_SKEPTIC": "<council seat: devil's advocate — failure modes and hidden assumptions>",
   "DESC_PRAGMATIST": "<council seat: weighs cost, effort, complexity vs payoff>",
@@ -42,7 +43,6 @@
   "DESC_COMMIT": "<one-line description: stage the right paths and write a focused commit>",
   "DESC_CODE_REVIEW": "<one-line description: review a diff/PR — correctness first, quality second>",
   "DESC_BRAINSTORM": "<one-line description: turn a raw idea into an approved design before code>",
-  "DESC_CLARIFY": "<one-line description: interview the user to pin goal, scope, success criteria>",
   "DESC_CODEBASE_DESIGN": "<one-line description: design deep modules behind small interfaces at deliberate seams>",
   "DESC_DOMAIN_MODELING": "<one-line description: maintain the domain glossary and record decisions as ADRs>",
   "DESC_WAYFINDER": "<one-line description: map an oversized effort as decision tickets and resolve them until the route is clear>",
@@ -57,19 +57,20 @@
   "DESC_PONYTAIL": "<one-line description: force the laziest working solution — minimal, shortest path, stdlib/native before custom code; lite/full/ultra intensity>",
   "DESC_PARALLEL_AGENTS": "<one-line description: fan independent subtasks to concurrent subagents, then converge>",
   "DESC_PIPELINE": "<one-line description: spawn a small crew — analyst, developer, tester — to work a substantial task in isolation, merging only on mechanical proof (tests + lint green)>",
-  "DESC_TDD": "<one-line description: drive implementation with a failing test first>",
   "DESC_DEVELOP": "<one-line description: drive a code change through cover-and-verify — smallest change, tests written for it, affected suite green, committed slice by slice>",
+  "DESC_TEACH": "<one-line description: teach toward a result — crash course, decode confusing material, or a day-by-day path to a deadline>",
+  "DESC_QUIZ": "<one-line description: test what the user knows instead of explaining — drill, explain-it-back check, or mastery claim stress-tested>",
+  "DESC_SECURITY_AUDIT": "<one-line description: review a change for how it would be attacked and what it leaks, ending in ship / fix / block>",
+  "DESC_SKILL_FORGE": "<one-line description: author, sharpen, merge, or retire a skill — reuse first, trigger tested, every satellite registered>",
+  "DESC_WORKTREE": "<one-line description: isolate a task in its own git worktree from a clean base, integrate through the normal gate, remove when merged>",
+  "DESC_CI_FIX": "<one-line description: turn a red CI run green at the cause — read the log, reproduce under CI's conditions, fix, watch the re-run>",
+  "DESC_DEPS_AUDIT": "<one-line description: survey dependencies — outdated, vulnerable, abandoned, unused — into a risk-ordered upgrade plan, installing nothing>",
+  "DESC_CONSOLIDATE_MEMORY": "<one-line description: tend the memory set — verify facts against the live repo, merge, retire, promote, rebuild the index>",
   "DESC_HANDOFF": "<one-line description: snapshot task state so a fresh context can resume in one read>",
   "DESC_SHIP": "<one-line description: open a PR with a structured body, or finish and merge a branch>",
   "DESC_RELEASE": "<one-line description: cut a release — bump, changelog, tag, kept in sync>",
   "DESC_INGEST": "<one-line description: convert a PDF/Office/web doc into clean markdown before reading>",
   "DESC_RESEARCH": "<one-line description: answer from the open web, verifying each claim against independent sources>",
-  "DESC_LEARNING_PATH": "<one-line description: turn a goal+deadline into a day-by-day path — daily timed task, correctness criterion, what not to do>",
-  "DESC_GAP_DETECTOR": "<one-line description: stress-test a mastery claim with five simple-looking questions, blunt verdict on the gaps>",
-  "DESC_FEYNMAN": "<one-line description: explain-it-simply check, caught on jargon/skipped steps/over-simplification, readout of what isn't solid>",
-  "DESC_CRASH_COURSE": "<one-line description: go from zero to functional in a skill fast — what to learn first, what to ignore, the one exercise that puts you ahead>",
-  "DESC_DRILL": "<one-line description: turn shaky knowledge into reflex — real situations, questioned through your mistakes, answer withheld until you've tried>",
-  "DESC_DECODE": "<one-line description: make confusing material click — one keystone sentence, an everyday analogy, then a comprehension check>",
   "DESC_HERDR": "<one-line description: drive herdr from inside it — split panes, spawn agents, read panes>",
   "DESC_RULE": "<one-line description: ask whether the user wants a standing rule or a durable memory, then write the memory, or pressure-test the rule until it is generic and adopt it in user-rules.md>",
   "DESC_PROFILE": "<one-line description: interview the user and draft or refresh PROFILE.md — who they are, how they work, how they like answers pitched; identity only, rules go to user-rules.md>",

package/themes/biker.json

@@ -23,6 +23,7 @@
   "DESC_DOCS": "Keeps the roadbook and the notes once the work is done.",
   "DESC_SECURITY": "Goes over anything touching the security surface, bolt by bolt.",
   "DESC_EXPLORER": "Sends the scout ahead — sweeps wide on a throwaway recon run, brings back only the read.",
+  "DESC_RESEARCHER": "Rides out past the county line — reads the road signs everyone else skipped, checks each one twice, brings back only what held up.",
   "DESC_ADVOCATE": "Argues for the run — the ground to be gained, the opening, the cost of staying parked.",
   "DESC_SKEPTIC": "Plays the doubter — calls the move on its failure modes, its risks, and the assumptions nobody checked.",
   "DESC_PRAGMATIST": "Weighs the run — the cost, the effort, the wear, and whether the payoff is worth the miles.",
@@ -42,7 +43,6 @@
   "DESC_COMMIT": "Load only what this run touched and seal a focused commit.",
   "DESC_CODE_REVIEW": "Go over a diff or PR — correctness first, finish after.",
   "DESC_BRAINSTORM": "Turn a raw idea into an agreed route before a single wheel turns.",
-  "DESC_CLARIFY": "Press the crew for the real goal, scope, and finish line of any run, then confirm and log the key calls before you roll out.",
   "DESC_CODEBASE_DESIGN": "Build deep modules — big behaviour packed behind a small handlebar, and place the seam on purpose.",
   "DESC_DOMAIN_MODELING": "Keep the crew's glossary alive in CONTEXT.md, and log every structural call as an ADR.",
   "DESC_WAYFINDER": "Chart a run too big for one ride as a shared map of decision tickets, cleared one at a time until the road's open.",
@@ -57,19 +57,20 @@
   "DESC_PONYTAIL": "The laziest fix that rides — least code, shortest road; stock parts before custom builds, one line before fifty. lite/full/ultra.",
   "DESC_PARALLEL_AGENTS": "Send several riders on independent runs at once, then bring in what they gather.",
   "DESC_PIPELINE": "Send a small crew of riders — scout, wrench, test rider — to work a job in their own lane, rolling it into the pack only once the road-test comes back clean.",
-  "DESC_TDD": "Let the test lead — set the failing test first, then only the code to pass it.",
   "DESC_DEVELOP": "Push a change through the run — the smallest turn of the wrench, a test written to cover it, the touched trials green, sealed slice by slice.",
+  "DESC_TEACH": "Get a rider road-ready — a crash course from never-ridden to rolling, a confusing manual decoded till it clicks, or a day-by-day route to a deadline.",
+  "DESC_QUIZ": "Test what they've actually got instead of explaining it — drill a move to reflex, make them talk you through it, or call the bluff on a claimed skill — with a straight readout of the gaps.",
+  "DESC_SECURITY_AUDIT": "Ride the change like someone trying to steal it — where it can be forced, what it leaks, what it drags in from outside — and call it: ship, fix, or park it.",
+  "DESC_SKILL_FORGE": "Forge, tune, merge, or scrap a move — reuse what's in the toolbox first, nail the trigger before the steps, test it against the neighbouring moves, register every bolt, doctor green.",
+  "DESC_WORKTREE": "Wrench on a task in its own bay — a fresh worktree off a clean base, nothing shared with the bike already on the lift, back through the normal gate, bay cleared when it's merged.",
+  "DESC_CI_FIX": "Get a red CI run rolling again at the cause — read the log, name the failure, reproduce under CI's own conditions, fix, watch the re-run.",
+  "DESC_DEPS_AUDIT": "Inspect every part on the bike — worn, recalled and reachable, discontinued, dead weight — into a risk-ordered replacement plan for migrate to execute, fitting nothing yet.",
+  "DESC_CONSOLIDATE_MEMORY": "Service the memory — check each note against the live machine, merge doubles, scrap stale ones, promote the lessons that keep biting to rules, rebuild the index.",
   "DESC_HANDOFF": "Hand off the bike — snapshot the state so a fresh rider takes the grips in one read.",
   "DESC_SHIP": "Roll it in: open a PR with a clear write-up, or merge the branch, once it's verified.",
   "DESC_RELEASE": "Cut a release — raise the version, write the roadbook entry, tag it, all three in sync.",
   "DESC_INGEST": "Bring a PDF, Office, or web document into clean markdown before reading it.",
   "DESC_RESEARCH": "Find the answer out on the open road — search wide, then prove every claim against independent sources before you trust it.",
-  "DESC_LEARNING_PATH": "Turn a goal and deadline into a day-by-day path — each day one timed task, a clear sign you nailed it, and what not to do.",
-  "DESC_GAP_DETECTOR": "Test a mastery claim — five simple-looking questions that expose shallow knowledge, with a blunt verdict on the gaps.",
-  "DESC_FEYNMAN": "Check it stuck — explain it plain, get pulled over on jargon, skipped steps, and over-simplification, then see what isn't solid.",
-  "DESC_CRASH_COURSE": "Get riding a skill fast — what to learn first, what to skip, and the one run that puts you ahead of most.",
-  "DESC_DRILL": "Beat shaky know-how into instinct — thrown into real situations and quizzed on your screw-ups, no answer 'til you've tried.",
-  "DESC_DECODE": "Cut through confusing material — the one line that matters first, a plain analogy, then a quick check before you roll on.",
   "DESC_HERDR": "Drive herdr from inside it — split panes, spawn sibling agents, wait on output, read other panes.",
   "DESC_RULE": "Ask first — standing rule, or just something to remember? Then log the memory, or lean on the rule till it holds every corner before bolting it into user-rules.md.",
   "DESC_PROFILE": "Stamp the rider's plate into PROFILE.md — who they are, what they ride, how they like the road called; identity only, club rules bolt into user-rules.md.",

package/themes/commentator.json

@@ -23,6 +23,7 @@
   "DESC_DOCS": "Files the race report and updates the notes once the chequered flag falls.",
   "DESC_SECURITY": "Runs the safety inspection on anything touching the security surface.",
   "DESC_EXPLORER": "Sends the track scout — sweeps wide from a throwaway recon lap, returns only the read.",
+  "DESC_RESEARCHER": "Sends the pit reporter down the paddock — checks every rumour against two sources before it goes on air.",
   "DESC_ADVOCATE": "Calls for the overtake — the place to be gained, the gap, the cost of staying put.",
   "DESC_SKEPTIC": "Plays the doubter — attacks the move for the failure modes, the risks, and the assumptions nobody checked on the grid.",
   "DESC_PRAGMATIST": "Weighs the strategy — the cost, the effort, the tyre wear, and whether the payoff is worth the stop.",
@@ -42,7 +43,6 @@
   "DESC_COMMIT": "Stage only the parts this lap touched and seal a focused commit.",
   "DESC_CODE_REVIEW": "Scrutineer a diff or PR — correctness first, polish after.",
   "DESC_BRAINSTORM": "Turn a raw idea into an agreed strategy before a single tyre hits the track.",
-  "DESC_CLARIFY": "Press the team for the real goal, scope, and the chequered-flag condition of any job, then confirm and log the key calls before lights out.",
   "DESC_CODEBASE_DESIGN": "Design deep modules — plenty of behaviour behind a tight interface, seam placed on purpose, not by accident.",
   "DESC_DOMAIN_MODELING": "Keep the paddock glossary alive in CONTEXT.md, and log every structural call as an ADR.",
   "DESC_WAYFINDER": "Chart work too big for one session as a shared race map of decision tickets, cleared one at a time until the way is clear.",
@@ -57,19 +57,20 @@
   "DESC_PONYTAIL": "The laziest line that finishes — least code, shortest path; standard kit before custom builds, one line before fifty. lite/full/ultra.",
   "DESC_PARALLEL_AGENTS": "Send several drivers out on independent stints at once, then bring in what they gather.",
   "DESC_PIPELINE": "Send a small crew — scout, mechanic, test driver — out on their own stint, and only wave them into the results once the timing sheet comes back clean.",
-  "DESC_TDD": "Let the trial lead the lap — set the failing test first, then only the code to pass it.",
   "DESC_DEVELOP": "Drive the change through cover-and-verify — the smallest setup tweak, a trial written to time it, the affected laps green, sealed corner by corner.",
+  "DESC_TEACH": "Bring a rookie up to speed — a crash course to race-ready, a confusing regulation decoded until it clicks, or a day-by-day programme to the deadline.",
+  "DESC_QUIZ": "Test the driver instead of briefing them — a drill to reflex, an explain-it-back debrief, or a claimed mastery stress-tested — with a frank readout of the gaps.",
+  "DESC_SECURITY_AUDIT": "Scrutineer a change for how it would be attacked and what it leaks — trust boundaries, the classic vulnerability classes, secrets, the supply chain — and rule: ship, fix, or black-flag.",
+  "DESC_SKILL_FORGE": "Author, sharpen, merge, or retire a manoeuvre — reuse first, the trigger before the steps, tested against its neighbours, every satellite registered, doctor green.",
+  "DESC_WORKTREE": "Run a task in its own garage — a fresh worktree off a clean base, nothing shared with the car already being worked on, back through the normal gate, garage cleared once merged.",
+  "DESC_CI_FIX": "Turn a red CI run green at the cause — read the telemetry, classify the failure, reproduce under CI's own conditions, fix, and watch the re-run.",
+  "DESC_DEPS_AUDIT": "Survey every component — outdated, on recall and reachable, discontinued, dead weight — into a risk-ordered upgrade plan for migrate to execute, fitting nothing yet.",
+  "DESC_CONSOLIDATE_MEMORY": "Tend the memory — verify every note against the live car, merge duplicates, retire stale ones, promote the lessons that keep recurring to rules, rebuild the index.",
   "DESC_HANDOFF": "Hand over the car — snapshot the state so a fresh driver takes the wheel in one read.",
   "DESC_SHIP": "Take the chequered flag: open a PR with a clear report, or merge the branch, once it's verified.",
   "DESC_RELEASE": "Cut a release — raise the version, write the race report, seal it with a tag, all three in sync.",
   "DESC_INGEST": "Bring a PDF, Office, or web document into clean markdown before reading it.",
   "DESC_RESEARCH": "Hunt the answer across the open circuit — search wide, then prove every claim against independent sources before you call it.",
-  "DESC_LEARNING_PATH": "Turn a goal and deadline into a day-by-day path — each day one timed task, a clear criterion, and what not to do.",
-  "DESC_GAP_DETECTOR": "Test a mastery claim live — five simple-looking questions that expose shallow knowledge, with a blunt verdict on the gaps.",
-  "DESC_FEYNMAN": "Check a topic stuck — explain it plain, get stopped on jargon, skipped steps, and over-simplification, then see what isn't solid.",
-  "DESC_CRASH_COURSE": "Up to speed in a skill fast — what to learn first, what to ignore, and the one move that puts you ahead of most.",
-  "DESC_DRILL": "Shaky form into reflex — out on the track, questioned through every mistake, the answer held until you've had a go.",
-  "DESC_DECODE": "Break a confusing section down — the one key line first, a plain analogy, then a check before we go on.",
   "DESC_HERDR": "Drive herdr from inside it — split panes, spawn sibling agents, wait on output, read other panes.",
   "DESC_RULE": "Ask first — standing house rule, or just one for the notebook? Then log it, or run the rule through the replay from every angle before it goes on the board in user-rules.md.",
   "DESC_PROFILE": "Build the player profile in PROFILE.md — position, form, how they like the commentary pitched; identity only, the rulebook stays in user-rules.md.",

package/themes/cyberpunk.json

@@ -23,6 +23,7 @@
   "DESC_DOCS": "Writes the user-facing data once the code goes live.",
   "DESC_SECURITY": "Breaks and checks ICE — audits anything on the security surface.",
   "DESC_EXPLORER": "Scouts the net — sweeps wide in a throwaway shell, returns only the intel.",
+  "DESC_RESEARCHER": "Jacks into the open net — pulls the intel, cross-checks it against two nodes, returns only what verified.",
   "DESC_ADVOCATE": "Pushes the build — the upside, the opening, the cost of sitting on it.",
   "DESC_SKEPTIC": "Runs devil's advocate — stress-tests the build for failure modes, risks, and buried assumptions.",
   "DESC_PRAGMATIST": "Weighs the spec — the cost, the effort, the complexity, and whether the payoff clears.",
@@ -42,7 +43,6 @@
   "DESC_COMMIT": "Stage only the sectors this run touched and push a focused commit.",
   "DESC_CODE_REVIEW": "Scan a diff or PR — correctness first, craft second.",
   "DESC_BRAINSTORM": "Turn a raw spec into an agreed design before any code hits the deck.",
-  "DESC_CLARIFY": "Jack into the user's real goal, scope, and win-condition for any job, then confirm and log the key calls before the run starts.",
   "DESC_CODEBASE_DESIGN": "Design deep modules — heavy behaviour behind a slim interface, seam placed on purpose, not left to chance.",
   "DESC_DOMAIN_MODELING": "Keep the domain glossary alive in CONTEXT.md, and log every structural call as an ADR.",
   "DESC_WAYFINDER": "Chart a run too big for one session as a shared map of decision tickets, cleared one at a time until the grid is clean.",
@@ -57,19 +57,20 @@
   "DESC_PONYTAIL": "The laziest build that runs — least code, shortest path; stock libs before custom rigs, one line before fifty. lite/full/ultra.",
   "DESC_PARALLEL_AGENTS": "Fork independent tasks to concurrent threads, then merge the output.",
   "DESC_PIPELINE": "Spin up a small crew of daemons — scout, coder, tester — on their own fork, and only merge the run once the trace comes back clean.",
-  "DESC_TDD": "Let the test drive — write the failing test first, then the minimum code to pass.",
   "DESC_DEVELOP": "Drive the change through the loop — smallest edit, a test written to cover it, the touched rigs run green, committed slice by slice.",
+  "DESC_TEACH": "Bring a user online — a crash course from zero to operational, confusing data decoded until it clicks, or a day-by-day route to a deadline.",
+  "DESC_QUIZ": "Probe what the user actually knows instead of explaining — drill to reflex, make them run it back, or stress-test a claimed mastery — with a cold readout of the gaps.",
+  "DESC_SECURITY_AUDIT": "Run the change like ICE would — where it can be breached, what it leaks, what it pulls in from outside the perimeter — and rule: ship, patch, or lock it down.",
+  "DESC_SKILL_FORGE": "Compile, patch, merge, or decommission a program — reuse the library first, trigger before routine, tested against neighbouring programs, every hook registered, diagnostics green.",
+  "DESC_WORKTREE": "Jack a task into its own sandbox — a fresh worktree off a clean base, nothing shared with the live rig, back through the normal gate, sandbox wiped once merged.",
+  "DESC_CI_FIX": "Turn a red CI run green at the cause — read the log, classify the failure, reproduce under CI's own conditions, patch, watch the re-run.",
+  "DESC_DEPS_AUDIT": "Scan every dependency on the rig — outdated, exploitable and reachable, abandoned, dead weight — into a risk-ordered upgrade plan for migrate, installing nothing yet.",
+  "DESC_CONSOLIDATE_MEMORY": "Defrag the datastore — verify every record against the live grid, merge duplicates, purge stale ones, promote recurring lessons to rules, rebuild the index.",
   "DESC_HANDOFF": "Pass the deck — dump the state so a fresh runner picks up in one read.",
   "DESC_SHIP": "Push it live: open a PR with a clear log, or merge the branch, once it's verified.",
   "DESC_RELEASE": "Cut a release — bump the version, write the changelog, tag it, all in sync.",
   "DESC_INGEST": "Rip a PDF, Office, or web document to clean markdown before reading it.",
   "DESC_RESEARCH": "Pull the answer off the net — scan wide, then cross-check every claim against independent sources before you trust it.",
-  "DESC_LEARNING_PATH": "Compile a goal and deadline into a day-by-day learning path — each day one timed task, a pass criterion, and what not to do.",
-  "DESC_GAP_DETECTOR": "Stress-test a mastery claim — five simple-looking queries that expose shallow knowledge, with a blunt verdict on the gaps.",
-  "DESC_FEYNMAN": "Verify a topic loaded — explain it simple, get flagged on jargon, skipped steps, and over-simplification, then see what isn't solid.",
-  "DESC_CRASH_COURSE": "Jack into a skill fast — what to load first, what to ignore, and the one run that puts you ahead of most.",
-  "DESC_DRILL": "Burn shaky knowledge into reflex — dropped into live situations and probed on every misfire, the answer locked until you've tried.",
-  "DESC_DECODE": "Decrypt garbled data — the one keystone line first, a plain analogy, then a three-question check before you jack out.",
   "DESC_HERDR": "Drive herdr from inside it — split panes, spawn sibling agents, wait on output, read other panes.",
   "DESC_RULE": "Ask first — standing rule, or just something to keep on file? Then write the memory, or stress the rule till it holds in every sector before flashing it into user-rules.md.",
   "DESC_PROFILE": "Compile the user's ident into PROFILE.md — who they are, their rig, the bandwidth they want answers at; identity only, hard directives flash to user-rules.md.",

package/themes/gamer.json

@@ -23,6 +23,7 @@
   "DESC_DOCS": "Writes the wiki and patch notes once the feature ships.",
   "DESC_SECURITY": "Runs the anti-cheat pass on anything touching the security surface.",
   "DESC_EXPLORER": "Scouts the map — sweeps wide in a throwaway scout run, returns only the intel.",
+  "DESC_RESEARCHER": "Reads the wiki and the patch notes — cross-checks every claim against two sources before it goes in the build guide.",
   "DESC_ADVOCATE": "Calls for the play — the upside, the opening, the cost of passing it up.",
   "DESC_SKEPTIC": "Runs the counter — attacks the play for failure modes, risks, and assumptions nobody checked.",
   "DESC_PRAGMATIST": "Weighs the build — the cost, the effort, the complexity, and whether the payoff's worth it.",
@@ -42,7 +43,6 @@
   "DESC_COMMIT": "Stage only what this run touched and save a focused commit.",
   "DESC_CODE_REVIEW": "Review a diff or PR — correctness first, polish second.",
   "DESC_BRAINSTORM": "Turn a raw idea into an agreed build before any code drops.",
-  "DESC_CLARIFY": "Quiz the player for the real goal, scope, and win-condition of any quest, then confirm and log the key picks before the run starts.",
   "DESC_CODEBASE_DESIGN": "Design deep modules — a lot of behaviour behind a small interface, seam placed on purpose, not farmed by accident.",
   "DESC_DOMAIN_MODELING": "Keep the domain glossary alive in CONTEXT.md, and log every structural call as an ADR entry.",
   "DESC_WAYFINDER": "Chart a quest too big for one run as a shared map of decision tickets, cleared one at a time until the fog lifts.",
@@ -57,19 +57,20 @@
   "DESC_PONYTAIL": "The laziest run that clears it — least code, shortest path; base kit before custom mods, one line before fifty. lite/full/ultra.",
   "DESC_PARALLEL_AGENTS": "Send the party on independent objectives at once, then regroup the loot.",
   "DESC_PIPELINE": "Send a small party — scout, builder, tester — into their own instance, and only merge the run once the clear is verified.",
-  "DESC_TDD": "Let the test lead — write the failing test first, then the minimum code to pass.",
   "DESC_DEVELOP": "Drive the change through the cover-and-verify loop — smallest move, tests written for it, the affected run green, committed slice by slice.",
+  "DESC_TEACH": "Level the player up toward a result — a crash course from tutorial to playable, a confusing mechanic decoded until it clicks, or a day-by-day grind to a deadline.",
+  "DESC_QUIZ": "Test the player instead of explaining — a drill to reflex, an explain-it-back check, or a claimed mastery stress-tested — with an honest readout of the gaps.",
+  "DESC_SECURITY_AUDIT": "Play the change like a griefer would — where it can be exploited, what it leaks, what it pulls in from outside — and call it: ship, patch, or block.",
+  "DESC_SKILL_FORGE": "Craft, tune, merge, or retire a combo — reuse the inventory first, trigger before inputs, tested against neighbouring combos, every satellite registered, doctor green.",
+  "DESC_WORKTREE": "Run a task in its own save slot — a fresh worktree off a clean base, nothing shared with the run in progress, back through the normal gate, slot deleted once merged.",
+  "DESC_CI_FIX": "Turn a red CI run green at the cause — read the log, classify the wipe, reproduce under CI's own conditions, fix, watch the re-run.",
+  "DESC_DEPS_AUDIT": "Audit the loadout — outdated, exploitable and reachable, abandoned, dead weight — into a risk-ordered upgrade plan for migrate to execute, equipping nothing yet.",
+  "DESC_CONSOLIDATE_MEMORY": "Tidy the save file — verify every note against the live game, merge duplicates, delete stale ones, promote recurring lessons to rules, rebuild the index.",
   "DESC_HANDOFF": "Pass the controller — snapshot the run so a fresh player picks up in one read.",
   "DESC_SHIP": "Submit the run: open a PR with clear notes, or merge the branch, once it's verified.",
   "DESC_RELEASE": "Ship a patch — bump the version, write the patch notes, tag it, all in sync.",
   "DESC_INGEST": "Load a PDF, Office, or web document into clean markdown before reading it.",
   "DESC_RESEARCH": "Look up the answer online — scout wide, then cross-check every claim against independent sources before you call it.",
-  "DESC_LEARNING_PATH": "Turn a goal and deadline into a day-by-day learning path — each day one timed quest, a clear win condition, and what not to do.",
-  "DESC_GAP_DETECTOR": "Boss-test a mastery claim — five simple-looking questions that expose shallow knowledge, with a blunt verdict on the gaps.",
-  "DESC_FEYNMAN": "Check a topic stuck — explain it simple, get stopped on jargon, skipped steps, and over-simplification, then see what isn't solid.",
-  "DESC_CRASH_COURSE": "Go from zero to playable in a skill fast — what to learn first, what to skip, and the one drill that puts you ahead of most.",
-  "DESC_DRILL": "Grind shaky knowledge into muscle memory — dropped into real fights and quizzed on your misplays, no answer until you've tried.",
-  "DESC_DECODE": "Make a confusing wall of text click — the one keystone line first, a plain analogy, then a check before you continue.",
   "DESC_HERDR": "Drive herdr from inside it — split panes, spawn sibling agents, wait on output, read other panes.",
   "DESC_RULE": "Ask first — standing house rule, or just something to remember? Then save the memory, or test the rule against its worst case before binding it into user-rules.md.",
   "DESC_PROFILE": "Fill in the player card in PROFILE.md — main role, loadout, preferred difficulty; identity only, house rules go in user-rules.md.",

package/themes/imperial.json

@@ -23,6 +23,7 @@
   "DESC_DOCS": "Inscribes the records for those who come after, once the work is done.",
   "DESC_SECURITY": "Wards the gates, auditing every change that touches the security surface.",
   "DESC_EXPLORER": "Ranges far and wide through a throwaway context, returning only the distilled truth.",
+  "DESC_RESEARCHER": "Ranges beyond the archives into the wider realm, holding every claim to two witnesses before it is entered as truth.",
   "DESC_ADVOCATE": "Champions the proposal — presses its strongest case, the prize to be won, the cost of staying the hand.",
   "DESC_SKEPTIC": "Stands as devil's advocate — assails the proposal for its weaknesses, its perils, and its unspoken assumptions.",
   "DESC_PRAGMATIST": "Weighs the campaign's true cost — the labour, the complexity, whether the spoils are worth the war.",
@@ -42,7 +43,6 @@
   "DESC_COMMIT": "Stage only what the deed touched, and seal it with a focused commit.",
   "DESC_CODE_REVIEW": "Scrutinise a diff or petition — correctness above all, craft thereafter.",
   "DESC_BRAINSTORM": "Forge a raw idea into an approved design before a single line of code is written.",
-  "DESC_CLARIFY": "Interrogate the objective to its true goal, scope, and measure of success, then confirm and inscribe the key decrees before the work is dispatched.",
   "DESC_CODEBASE_DESIGN": "Design deep modules — vast behaviour behind a narrow interface, the seam placed by decree, not by chance.",
   "DESC_DOMAIN_MODELING": "Keep the domain glossary alive in CONTEXT.md, and seal each structural decree as an ADR.",
   "DESC_WAYFINDER": "Chart a campaign too vast for one session as a shared map of decision decrees, resolved one at a time until the way is clear.",
@@ -57,19 +57,20 @@
   "DESC_PONYTAIL": "Compel the most frugal working rite — minimal and shortest-path; the standard armoury before bespoke code, one line before fifty. lite/full/ultra.",
   "DESC_PARALLEL_AGENTS": "Loose many specialists upon independent tasks at once, then gather what they bring.",
   "DESC_PIPELINE": "Dispatch a small cohort of Adepts — scout, artisan, examiner — to labour in their own sealed chamber, sanctioned into the Reliquary only once the proving-rite returns clean.",
-  "DESC_TDD": "Let the trials lead the work — set the failing trial first, then only the code to pass it.",
   "DESC_DEVELOP": "Drive a change through the cover-and-verify loop — the smallest stroke, the trials written to cover it, the affected suite green, sealed slice by slice.",
+  "DESC_TEACH": "Instruct the supplicant toward a result — a crash course from ignorance to function, a confusing text decoded until it clicks, or a day-by-day path to the appointed hour.",
+  "DESC_QUIZ": "Test what the supplicant knows rather than teach it — a drill to reflex, a recitation caught on every stumble, or a claimed mastery put to the ordeal — with an unsparing readout of the gaps.",
+  "DESC_SECURITY_AUDIT": "Scrutinise a change as the enemy would — where it may be breached, what it leaks, what it admits from beyond the walls — and pronounce: ship, purge, or bar it.",
+  "DESC_SKILL_FORGE": "Forge, sharpen, merge, or retire a rite — reuse the canon first, the trigger before the steps, tested against its neighbouring rites, every satellite inscribed, doctor green.",
+  "DESC_WORKTREE": "Wall a task off in its own cell — a fresh worktree from a clean base, nothing shared with the work already at the bench, returned through the proper gate, the cell razed once merged.",
+  "DESC_CI_FIX": "Turn a red CI run green at the cause — read the log, classify the failure, reproduce under CI's own conditions, mend, and watch the re-run.",
+  "DESC_DEPS_AUDIT": "Survey every dependency of the machine — outdated, corrupted and reachable, abandoned, dead weight — into a risk-ordered plan for migrate to execute, installing nothing yet.",
+  "DESC_CONSOLIDATE_MEMORY": "Tend the memory — verify each record against the living archive, merge duplicates, retire the stale, promote the lessons that recur to law, rebuild the index.",
   "DESC_HANDOFF": "Inscribe the state of the work so a fresh mind may take up the watch without loss.",
   "DESC_SHIP": "Send the finished work to the gate — a pull request well-formed, or the branch merged once the trials are passed.",
   "DESC_RELEASE": "Proclaim a new version — raise the number, inscribe the chronicle, seal it with a tag, all three in accord.",
   "DESC_INGEST": "Render a foreign document — PDF, Office, or web — into clean markdown before it may be read.",
   "DESC_RESEARCH": "Seek the answer from the wider world — range across many sources, then prove each claim against independent witnesses before you trust it.",
-  "DESC_LEARNING_PATH": "Forge a concrete goal and deadline into a day-by-day path of study — each day one timed task, a measure of correctness, and a forbidden misstep.",
-  "DESC_GAP_DETECTOR": "Put a claim of mastery to the question — five deceptively simple trials that expose shallow understanding, and a blunt verdict on what is wanting.",
-  "DESC_FEYNMAN": "Prove whether the lesson took — explain it in plain speech, be halted at every jargon, skipped step, or false simplification, then face what is not yet solid.",
-  "DESC_CRASH_COURSE": "Forge a recruit functional in a skill with haste — what to master first, what to ignore, and the one drill that sets you ahead of most.",
-  "DESC_DRILL": "Temper shaky knowledge into reflex — cast into real situations and questioned through your failures, the answer withheld until you have tried.",
-  "DESC_DECODE": "Lay impenetrable lore bare — the one keystone line first, a plain parable, then a trial before you advance.",
   "DESC_HERDR": "Drive herdr from inside it — split panes, spawn sibling agents, wait on output, read other panes.",
   "DESC_RULE": "Ask first — a standing edict, or a matter merely to be remembered? Then inscribe the memory, or try the edict against the case that would break it before it enters user-rules.md.",
   "DESC_PROFILE": "Inscribe the user's identity into PROFILE.md — grant audience, ask who they are and how they will to be served; identity only, edicts belong in user-rules.md.",