Add /dig steering skill

docs/index.html

@@ -574,12 +574,24 @@ <h1>stbenjam&#x27;s skills</h1>
           "description_html": "Make the response terser while preserving important reasoning, decisions, and next steps.",
           "meta": "License: MIT"
         },
+        {
+          "name": "dig",
+          "description": "Revisit superficial or under-verified agent work by investigating its claims, assumptions, evidence, and missing edge cases.",
+          "description_html": "Revisit superficial or under-verified agent work by investigating its claims, assumptions, evidence, and missing edge cases.",
+          "meta": "License: MIT"
+        },
         {
           "name": "lfg",
           "description": "Implement the requested direction immediately using reasonable defaults and no unnecessary follow-up questions.",
           "description_html": "Implement the requested direction immediately using reasonable defaults and no unnecessary follow-up questions.",
           "meta": "License: MIT"
         },
+        {
+          "name": "map",
+          "description": "Explain the shape of an implementation, its key architectural decisions, and how its main pieces fit together.",
+          "description_html": "Explain the shape of an implementation, its key architectural decisions, and how its main pieces fit together.",
+          "meta": "License: MIT"
+        },
         {
           "name": "nah",
           "description": "Abandon the current direction, propose three genuinely different approaches, and recommend the strongest one.",

plugins/steering/README.md

@@ -11,5 +11,7 @@ Code, Codex, and standalone Agent Skills setup.
 - [`/wow`](skills/wow/SKILL.md) — raise a frontend redesign to an award-worthy bar
 - [`/cut`](skills/cut/SKILL.md) — make the response terser without losing substance
 - [`/ask`](skills/ask/SKILL.md) — pause and ask the questions needed for a high-quality solution
+- [`/dig`](skills/dig/SKILL.md) — revisit superficial or under-verified agent work
+- [`/map`](skills/map/SKILL.md) — explain the shape and architecture of an implementation
 
 Inspired by [`bro` from dmmulroy/skills](https://github.com/dmmulroy/skills/blob/main/bro/SKILL.md).

plugins/steering/skills/dig/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: dig
+description: Revisit superficial or under-verified agent work by investigating its claims, assumptions, evidence, and missing edge cases.
+license: MIT
+metadata:
+  author: stbenjam
+  version: "0.1.0"
+user-invocable: true
+disable-model-invocation: true
+---
+
+Revisit the preceding work skeptically: identify shallow reasoning, unsupported claims, missing edge cases, and unverified assumptions, then investigate the highest-risk gaps.

plugins/steering/skills/dig/agents/openai.yaml

@@ -0,0 +1,10 @@
+interface:
+  display_name: "Dig"
+  short_description: "Dig deeper into superficial agent work."
+  icon_small: "./assets/icon.svg"
+  icon_large: "./assets/icon.svg"
+  brand_color: "#0E7490"
+  default_prompt: "Use $dig to investigate superficial agent work."
+
+policy:
+  allow_implicit_invocation: false

plugins/steering/skills/map/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: map
+description: Explain the shape of an implementation, its key architectural decisions, and how its main pieces fit together.
+license: MIT
+metadata:
+  author: stbenjam
+  version: "0.1.0"
+user-invocable: true
+disable-model-invocation: true
+---
+
+Describe the broad strokes of the preceding implementation and provide a simple map of its shape. Walk through the key architectural decisions, how the main pieces fit together, and the most important tradeoffs.

plugins/steering/skills/map/agents/openai.yaml

@@ -0,0 +1,10 @@
+interface:
+  display_name: "Map"
+  short_description: "Explain an implementation's shape and architecture."
+  icon_small: "./assets/icon.svg"
+  icon_large: "./assets/icon.svg"
+  brand_color: "#4338CA"
+  default_prompt: "Use $map to explain the shape of an implementation."
+
+policy:
+  allow_implicit_invocation: false

skills/dig

@@ -0,0 +1 @@
+../plugins/steering/skills/dig

skills/map

@@ -0,0 +1 @@
+../plugins/steering/skills/map