Merge pull request #461 from mattpocock/fix/wayfinder-self-grilling

wayfinder/grilling: stop the agent grilling itself

.changeset/negation-failure-mode.md

@@ -0,0 +1,5 @@
+---
+"mattpocock-skills": minor
+---
+
+Add two adjacent Steering failure modes to **`writing-great-skills`**, both about how language you think of as "off" still steers the agent. **Negation** — the _elephant_ — is steering by prohibition: naming what _not_ to do drags the forbidden behaviour into context and makes it _more_ available, not less (_don't think of an elephant_), so the cure is to prompt the **positive**. **Negative Space** — the void — is blindness to the steering done by what you leave _out_: every decision a skill declines is delegated to the agent's priors rather than left neutral, so the cure is to read a draft for its silences and decide each omission deliberately (fill it, or leave it open as a real **branch**). Kept as two entries, not one — they carry different diagnostics and different cures — each a full `GLOSSARY.md` entry plus a `SKILL.md` failure-mode bullet, matching how every other failure mode is carried.

.changeset/wayfinder-grilling-not-self.md

@@ -0,0 +1,12 @@
+---
+"mattpocock-skills": patch
+---
+
+Stop **`wayfinder`** grilling *itself* instead of the human, and classify ticket types by **HITL** / **AFK**.
+
+Students reported `/wayfinder` picking a **grilling** ticket and answering its own questions rather than turning to the human. Root cause: **`grilling`** carried a line written for the live-human case — "If a question can be answered by exploring the codebase, explore the codebase instead" — which, once `wayfinder` runs grilling inside a resolve-the-ticket frame, reads as license to answer questions autonomously and race the ticket to "resolved".
+
+Two changes:
+
+- **`grilling`** now splits *facts* (look them up) from *decisions* (put each one to the human and wait for their answer).
+- **`wayfinder`** labels every ticket type **HITL** (human in the loop — Prototype, Grilling) or **AFK** (agent alone — Research; Task is either). A HITL ticket only resolves through the live exchange, so the "wait for the human" behaviour falls out of the label instead of being spelled out per type — a grilling agent that answers its own questions has, by definition, broken HITL.

skills/in-progress/wayfinder/SKILL.md

@@ -71,10 +71,12 @@ The answer isn't part of the body — it's recorded on resolution (see [Work thr
 
 ## Ticket Types
 
-- **Research**: Reading documentation, third-party APIs, or local resources like knowledge bases. Creates a markdown summary as a linked asset. Use when knowledge outside the current working directory is required.
+Every ticket is either **HITL** — human in the loop, worked *with* a human who speaks for themselves — or **AFK**, driven by the agent alone. A HITL ticket only resolves through that live exchange; the agent never stands in for the human's side of it (a grilling agent that answers its own questions has broken this).
-- **Prototype**: Raise the fidelity of the discussion by making a cheap, rough, concrete artifact to react to — an outline, a rough take, a stub, or UI/logic code via the /prototype skill. Links the prototype as an asset. Use when "how should it look" or "how should it behave" is the key question.
+
-- **Grilling**: Conversation with the agent. Uses the /grilling and /domain-modeling skills. Asks one question at a time. The default case.
+- **Research** (AFK): Reading documentation, third-party APIs, or local resources like knowledge bases. Creates a markdown summary as a linked asset. Use when knowledge outside the current working directory is required.
-- **Task**: Manual work that must happen before a *decision* can be made — nothing to decide, prototype, or research, but the discussion is blocked until it's done. Signing up for a service so its API can be judged, provisioning access, moving data so its shape can be seen. This is the one type that *does* rather than decides — and it earns its place by unblocking a decision, not by delivering the destination. The agent automates it where it can; otherwise it hands the human a precise checklist. Resolved when the work is done; the answer records what was done and any resulting facts (credentials location, new URLs, row counts) later tickets depend on.
+- **Prototype** (HITL): Raise the fidelity of the discussion by making a cheap, rough, concrete artifact to react to — an outline, a rough take, a stub, or UI/logic code via the /prototype skill. Links the prototype as an asset. Use when "how should it look" or "how should it behave" is the key question.
+- **Grilling** (HITL): Conversation via the /grilling and /domain-modeling skills, one question at a time. The default case.
+- **Task** (HITL or AFK): Manual work that must happen before a *decision* can be made — nothing to decide, prototype, or research, but the discussion is blocked until it's done. Signing up for a service so its API can be judged, provisioning access, moving data so its shape can be seen. This is the one type that *does* rather than decides — and it earns its place by unblocking a decision, not by delivering the destination. The agent drives it alone where it can (AFK); otherwise it hands the human a precise checklist (HITL). Resolved when the work is done; the answer records what was done and any resulting facts (credentials location, new URLs, row counts) later tickets depend on.
 
 ## Fog of war
 

skills/productivity/grilling/SKILL.md

@@ -7,6 +7,6 @@ Interview me relentlessly about every aspect of this plan until we reach a share
 
 Ask the questions one at a time, waiting for feedback on each question before continuing. Asking multiple questions at once is bewildering.
 
-If a question can be answered by exploring the codebase, explore the codebase instead.
+If a *fact* can be found by exploring the codebase, look it up rather than asking me. The *decisions*, though, are mine — put each one to me and wait for my answer.
 
 Do not enact the plan until I confirm we have reached a shared understanding.

skills/productivity/writing-great-skills/GLOSSARY.md

@@ -158,6 +158,12 @@ _Failure mode._ Ending the current step before it is genuinely done, because the
 
 _Avoid_: premature closure, the rush, rushing, shortcutting
 
+### Negation
+
+_Failure mode._ Steering by prohibition — telling the agent what _not_ to do — which drags the forbidden behaviour into context and makes it _more_ available, not less. _Don't think of an elephant_, and the elephant is all there is; _never write verbose comments_, and verbosity is the pattern the agent has just read. The negation is a weak modifier the strongly-activated concept overruns, so the ban half-reads as an instruction to do the thing. Its **leading word** is the _elephant_: whatever a prohibition names into the frame. Cure: prompt the **positive** — describe the target behaviour ("write one-line comments") so the banned one is never spoken. A prohibition earns its place only as a hard guardrail on a behaviour you cannot phrase positively; even then, pair it with the positive target so attention lands on what to do.
+
+_Avoid_: ironic rebound, don't-prompting, the pink elephant
+
 ## Pruning
 
 Keeping a skill lean — each remedy paired with the failure it cures.

skills/productivity/writing-great-skills/SKILL.md

@@ -80,3 +80,4 @@ Use these to diagnose issues the user may be having with the skill.
 - **Sediment** — stale layers that settle because adding feels safe and removing feels risky. The default fate of any skill without a pruning discipline.
 - **Sprawl** — a skill simply too long, even when every line is live and unique. Hurts readability and maintainability and wastes tokens. The cure is the ladder: disclose **reference** behind pointers, and split by **branch** or sequence so each path carries only what it needs.
 - **No-op** — a line the model already obeys by default, so you pay load to say nothing. The test: does it change behaviour versus the default? A weak leading word (_be thorough_ when the agent is already thorough-ish) is a no-op; the fix is a stronger word (_relentless_), not a different technique.
+- **Negation** — steering by prohibition backfires: _don't think of an elephant_ names the elephant and makes it more available, not less. Prompt the **positive** — state the target behaviour so the banned one is never spoken; keep a prohibition only as a hard guardrail you can't phrase positively, and even then pair it with what to do instead.