From 80f29a4f29a6df088020916f30453d5e1a9cf6f5 Mon Sep 17 00:00:00 2001 From: Matt Pocock Date: Mon, 6 Jul 2026 13:49:33 +0100 Subject: [PATCH 1/2] wayfinder: reassert planning-only purpose with a leading-word section Add a "Plan, don't do" section as the single source of truth for the plan/execute boundary, and bring two passages into line with it: - Destination is always a set of resolved decisions, never the executed change itself (the migration example previously read as execution). - The Task ticket type now earns its place only by unblocking a decision, pointing back to the new section rather than restating the principle. Co-Authored-By: Claude Opus 4.8 (1M context) --- skills/in-progress/wayfinder/SKILL.md | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/skills/in-progress/wayfinder/SKILL.md b/skills/in-progress/wayfinder/SKILL.md index 5e1b6cb..3c08745 100644 --- a/skills/in-progress/wayfinder/SKILL.md +++ b/skills/in-progress/wayfinder/SKILL.md @@ -5,7 +5,13 @@ description: Plan a huge chunk of work — more than one agent session can hold A loose idea has arrived — too big for one agent session, and wrapped in fog: the way from here to the **destination** isn't visible yet. Wayfinding is about finding that way, not charging at the destination. This skill charts the way as a **shared map** on the repo's issue tracker, then works its tickets one at a time until the route is clear. -The destination varies per effort, and naming it is the first act of charting — it shapes every ticket. It might be a spec to hand off and iterate on, a decision to lock before planning starts, or a change made in place like a data-structure migration. The map is domain-agnostic — engineering work, course content, whatever fits the shape. +The destination varies per effort, and naming it is the first act of charting — it shapes every ticket. It might be a spec to hand off and iterate on, a single decision locked before planning starts, or a plan of attack for a change like a data-structure migration — but always a set of **resolved decisions**, never the executed change itself. The map is domain-agnostic — engineering work, course content, whatever fits the shape. + +## Plan, don't do + +Wayfinder is **planning**, not **execution**. Every ticket resolves a decision; the map is done when the way is clear — the point where someone could *go and do the thing* with nothing left to decide. The doing is a separate act, after the map closes. A session that starts building the destination — writing the migration, shipping the feature — has stopped wayfinding, and should stop. + +So the map's output is decisions, not deliverables. When you feel the pull to just do the work, that pull is the signal you've reached the edge of the map: the way is clear, and it's time to hand off. ## Refer by name @@ -70,7 +76,7 @@ The answer isn't part of the body — it's recorded on resolution (see [Work thr - **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. - **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. -- **Task**: Literal manual work that must be done before the discussion can move forward — nothing to decide, prototype, or research. Moving data, signing up for a service, provisioning access. 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. +- **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 only by unblocking a decision — never by delivering the destination (that's execution; see [Plan, don't do](#plan-dont-do)). 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. ## Fog of war From 7be28cf5e64a7c990e42db44ea97665df2adfa75 Mon Sep 17 00:00:00 2001 From: Matt Pocock Date: Mon, 6 Jul 2026 13:53:59 +0100 Subject: [PATCH 2/2] wayfinder: planning by default, overridable via Notes; shrink section Address review: don't preclude execution. Wayfinder is planning by default, but an effort's Notes can carry execution into the map. Collapse the section to one paragraph, revert the destination sentence to its original (which already allowed an executed change), and drop the Task cross-reference to match. Co-Authored-By: Claude Opus 4.8 (1M context) --- skills/in-progress/wayfinder/SKILL.md | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/skills/in-progress/wayfinder/SKILL.md b/skills/in-progress/wayfinder/SKILL.md index 3c08745..d7854ef 100644 --- a/skills/in-progress/wayfinder/SKILL.md +++ b/skills/in-progress/wayfinder/SKILL.md @@ -5,13 +5,11 @@ description: Plan a huge chunk of work — more than one agent session can hold A loose idea has arrived — too big for one agent session, and wrapped in fog: the way from here to the **destination** isn't visible yet. Wayfinding is about finding that way, not charging at the destination. This skill charts the way as a **shared map** on the repo's issue tracker, then works its tickets one at a time until the route is clear. -The destination varies per effort, and naming it is the first act of charting — it shapes every ticket. It might be a spec to hand off and iterate on, a single decision locked before planning starts, or a plan of attack for a change like a data-structure migration — but always a set of **resolved decisions**, never the executed change itself. The map is domain-agnostic — engineering work, course content, whatever fits the shape. +The destination varies per effort, and naming it is the first act of charting — it shapes every ticket. It might be a spec to hand off and iterate on, a decision to lock before planning starts, or a change made in place like a data-structure migration. The map is domain-agnostic — engineering work, course content, whatever fits the shape. ## Plan, don't do -Wayfinder is **planning**, not **execution**. Every ticket resolves a decision; the map is done when the way is clear — the point where someone could *go and do the thing* with nothing left to decide. The doing is a separate act, after the map closes. A session that starts building the destination — writing the migration, shipping the feature — has stopped wayfinding, and should stop. - -So the map's output is decisions, not deliverables. When you feel the pull to just do the work, that pull is the signal you've reached the edge of the map: the way is clear, and it's time to hand off. +Wayfinder is **planning** by default: each ticket resolves a decision, and the map is done when the way is clear — nothing left to decide before someone goes and does the thing. The pull to just do the work is usually the signal you've reached the edge of the map and it's time to hand off. An effort can override this in its **Notes** — carrying execution into the map itself — but absent that, produce decisions, not deliverables. ## Refer by name @@ -76,7 +74,7 @@ The answer isn't part of the body — it's recorded on resolution (see [Work thr - **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. - **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. -- **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 only by unblocking a decision — never by delivering the destination (that's execution; see [Plan, don't do](#plan-dont-do)). 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. +- **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. ## Fog of war