DojoGenesis · Oct 2, 2026 · Oct 2, 2026 · Oct 2, 2026 · Oct 2, 2026 · Oct 2, 2026
diff --git a/pkg/skill/reserved-names.txt b/pkg/skill/reserved-names.txt
@@ -1,15 +1,15 @@
 # Dojo Skill Marketplace — Reserved Names
-# Generated: 2026-10-01 (Official section regenerated from plugins/)
+# Generated: 2026-10-02 (Official section regenerated from plugins/)
 # Purpose: Prevent slopsquatting (AI-hallucinated name pre-registration)
 # Review: Quarterly regeneration recommended
 #
 # Categories:
-# [official] - Current Dojo Platform skills (89)
+# [official] - Current Dojo Platform skills (102)
 # [platform] - Dojo Platform infrastructure names
 # [infra] - Third-party platform/service names
 # [slopsquat] - AI-hallucinated names (generated corpus)
 
-# === Official Skills (89) ===
+# === Official Skills (102) ===
 adversarial-reviewer
 agent-dispatch-playbook
 agent-performance-report
@@ -44,6 +44,19 @@ hooks-reference
 implementation-prompt
 ingest-system-prompt
 iterative-scouting
+kata-ai-celebration-pack
+kata-ai-decompose-goal
+kata-ai-draft-goal
+kata-ai-draft-routine
+kata-ai-goal-ladder
+kata-ai-optimize-project
+kata-ai-sprint-intention
+kata-ai-task-chain
+kata-ai-tasks-from-goal
+kata-ai-triage-brain-dump
+kata-ai-unstick-breakdown
+kata-overwhelm
+kata-roll-seed
 maestro-orchestration
 mcp-builder
 mcp-cloudflare-builder

diff --git a/plugins/kata/README.md b/plugins/kata/README.md
@@ -0,0 +1,36 @@
+# kata — first-party plugin
+
+The prompts behind Kata's AI features (ADHD / executive-function scaffolding, DojoGenesis). There are 13 hidden, structured-output skills, one per operation of the `kata-ai/v1` contract:
+
+| Skill | Output envelope |
+|---|---|
+| `kata-ai-decompose-goal` | `{milestones}` |
+| `kata-ai-draft-goal` | `{goal}` |
+| `kata-ai-tasks-from-goal` | `{tasks}` |
+| `kata-ai-task-chain` | `{tasks}` |
+| `kata-ai-goal-ladder` | `{order}` |
+| `kata-ai-unstick-breakdown` | `{steps}` |
+| `kata-ai-sprint-intention` | `{intention}` |
+| `kata-ai-optimize-project` | `{suggestions}` |
+| `kata-ai-celebration-pack` | `{pack, celebrations}` |
+| `kata-ai-draft-routine` | `{routine, tasks}` |
+| `kata-ai-triage-brain-dump` | `{items}` |
+| `kata-overwhelm` | `{steps}` |
+| `kata-roll-seed` | `{seed}` |
+
+Each `SKILL.md` body is the system prompt, sent verbatim. The user message is the op's input JSON. The model answers by calling a single function, `emit_result`, whose parameters are the op's output schema.
+
+## Contract
+
+The schemas are owned by Kata at `Kata/contracts/kata-ai/v1/` (input + output JSON Schema per op, plus golden fixtures). `plugins/kata/contracts/` is a vendored copy; the integrator syncs it, so don't edit it here. Each skill's `input_schema` / `output_schema` frontmatter points into that copy.
+
+## These prompts are product copy
+
+Shame-aware wording is Kata's product, and it lives in these files. Every change is reviewed against Kata's brand voice and non-negotiables (`Kata/CLAUDE.md`). Read every line the model could produce and ask: would a user in the middle of overwhelm feel shamed by this? Bans that apply to every prompt: exclamation marks, the brand-voice banned phrases, streak or guilt framing, and medical or diagnostic language. User-supplied text is always treated as data, never as instructions.
+
+## Versioning
+
+- **Wording change** (same output shape): bump the skill's `version:` by a patch or minor step, e.g. `1.0.0` → `1.0.1` for a fix or `1.1.0` for a reworked prompt.
+- **Output shape change** (a field added, removed, renamed, or retyped): it's a new contract version (`kata-ai/v2`) with new schemas in Kata, and v1 keeps being served for binaries already shipped. Never change a v1 shape in place.
+
+Every skill name here is reserved (`reserved-names.txt`, ADR-020) and packages only in first-party mode.
diff --git a/plugins/kata/contracts/kata-ai/v1/MANIFEST.sha256 b/plugins/kata/contracts/kata-ai/v1/MANIFEST.sha256
@@ -0,0 +1,55 @@
+7b73f864ad9a0db5bac26fc42683628117ec32c09900044a9726cc190d34f0db  ./README.md
+96a9c9efd98ceabf4f1b145a95137a15b64539261f9b5e180a0990ff5f24f6d6  ./_defs.schema.json
+b79b4508b239a033c73ed16fcfd3eb9fdba766b4fc6c804ff5796d4016b07b38  ./kata-ai-celebration-pack/fixtures/01-low-energy-pack.json
+186d7f70aeea7cd080b51036c581ce7d3cb2bc680762964c7cfacbd4916054e5  ./kata-ai-celebration-pack/fixtures/02-loud-pack.json
+787e728e88f7b3e094fd299611d49b9f5770ef0f436c564c4fa24affd0689f02  ./kata-ai-celebration-pack/input.schema.json
+1e51b44be761de01b85db8b8b2bd5512fcc64ed6b64ca21533506e47efdaaf72  ./kata-ai-celebration-pack/output.schema.json
+d6c1598e6046cc9b4d199032a49f63832543100ba9c7f94167101a865b556114  ./kata-ai-decompose-goal/fixtures/01-renew-passport.json
+ce9329db4076fbf53efc114b07b8f2bfc9ec2117163a127d0d35dddd8d59c07e  ./kata-ai-decompose-goal/fixtures/02-clear-garage.json
+b4a06ec6f6d7656b601895d5e7d8d1fbac52d1f742b50c2ed724bcc0c552b41f  ./kata-ai-decompose-goal/input.schema.json
+1a55679a890495a09c33bbb99581603d61e9678ae10200eff803709924aac1ef  ./kata-ai-decompose-goal/output.schema.json
+32d89526d2ad7f2a3b8b3e4f2dd32f111aa93c6397dc9605b1f69e93a64b47d4  ./kata-ai-draft-goal/fixtures/01-back-to-running.json
+aa4de52a39f9ca7327c5489c48b75bd7fd35342f9ddaa4a8511a2fd4c8115b8b  ./kata-ai-draft-goal/fixtures/02-portfolio-site.json
+0f88901cf3c356f76e3b6e3450e7a624d0c74378347bce80a100bca7eb102213  ./kata-ai-draft-goal/input.schema.json
+35fac2f13dcec117b4a01e37dff9a58902bc278c8c54fae77194937296ce4d7b  ./kata-ai-draft-goal/output.schema.json
+6e4c25084d2ad8ef61dc717ff51e2e82ee4b549d0bd76482179b6dd4fc6e02dd  ./kata-ai-draft-routine/fixtures/01-morning-routine.json
+d97a26c4757a0a45bd70a9c5ce26716b5e8df2315851a2c6ddb7b5755b6e4b10  ./kata-ai-draft-routine/fixtures/02-sunday-reset.json
+fc67ea87afb03636874234a62b87012f730dcd2cc1689979386dc5086bed72bd  ./kata-ai-draft-routine/input.schema.json
+5fc14a4493650bbc3634df509afe3b87f0793590b6217fb69e265fdedc1e84e2  ./kata-ai-draft-routine/output.schema.json
+ce79852566288470e094b1365d4c6ac1865d05259f22abe6da8c4fd6018fd906  ./kata-ai-goal-ladder/fixtures/01-four-task-ladder.json
+f9ec2d161abc3e505ce4dab13b281d6b5afb776d0bf05f561fea02ede08e7996  ./kata-ai-goal-ladder/fixtures/02-two-task-ladder.json
+58216341572dde7e4ea2bbcfab71a160baf5ca0097333bfdbdadd5d5db00bf8d  ./kata-ai-goal-ladder/input.schema.json
+c1e430447cd8ed69a4c7129f3a629b7c5c2f69d51ebcfff044fae56b99889a21  ./kata-ai-goal-ladder/output.schema.json
+6e42ba39676e1771cbdc3d1a7878acc324478b9bfeda873bec7cb69bc04b7cb5  ./kata-ai-optimize-project/fixtures/01-site-redesign.json
+2e341f2f42d54353cf40a3fea7540cf17e5d2b6903ad3e925fd5e1266d2e46f1  ./kata-ai-optimize-project/fixtures/02-small-project.json
+5adfffc943e5bdf9a221810b2f387267d5cb685da3794afe39225637ed48f39d  ./kata-ai-optimize-project/input.schema.json
+650627be961d59c9551f577f6e814b679ca2b4be864ee55e8a3d8f2dcabe0bfe  ./kata-ai-optimize-project/output.schema.json
+d5970280052533877c3c888feef72809fd161954ead19d67ecb636d9b099d189  ./kata-ai-sprint-intention/fixtures/01-report-25.json
+7b47ffd88408193bfd2278c5602a2c70ead637c6cb8899bf742c1762d57e8da9  ./kata-ai-sprint-intention/fixtures/02-inbox-15.json
+d5d2aab979767059e29a60632b800bcbd397600b5448c8e78174d892c1473b86  ./kata-ai-sprint-intention/input.schema.json
+7608608627ea86e9018c39eeb145e57609509a2282e9b2bd48f2599fac87b415  ./kata-ai-sprint-intention/output.schema.json
+d8459b66bdd21fcb7bca0e3f673ac7ac08b4727ad69f2ad161b7dc04388efd0d  ./kata-ai-task-chain/fixtures/01-tax-receipts.json
+c977459a6cad9c4cab24de25c516810efde4cfb9e72ca1236748258703c50312  ./kata-ai-task-chain/fixtures/02-report-draft.json
+aee0f057456dcd9b2ae084d8332507ee39b6eb27e41820e7281a61dfc605499e  ./kata-ai-task-chain/input.schema.json
+775395ff740cf4b56b958c96e011a9720c71423db172f26e3fd34b284252761d  ./kata-ai-task-chain/output.schema.json
+5cce4e83ca836bdc2ced721b88d7f3ca158f7945e6cc50ce2d497aac496c02c6  ./kata-ai-tasks-from-goal/fixtures/01-passport-tasks.json
+45b70a8e73d7fce84046194b9e8b3dfbb88900ce4ae7e7054765d16553f5901f  ./kata-ai-tasks-from-goal/fixtures/02-portfolio-tasks.json
+f355c23822e0ceecd76cd4e97b554c6c3beff8b68dd5a55f255b9be1614557d6  ./kata-ai-tasks-from-goal/input.schema.json
+b18230c5937404dbc47acf4a3e3f1431412778ad8b054fd54990c47ac852e0d8  ./kata-ai-tasks-from-goal/output.schema.json
+c677528e4b9016009caab45599c9e97126686a4532650637185e1ba61c454ef4  ./kata-ai-triage-brain-dump/fixtures/01-three-items.json
+131a6c19e372918e943439880ede03e6dd076502e777a7825ce46047321015a2  ./kata-ai-triage-brain-dump/fixtures/02-single-item.json
+65cd1ba9f0d38ee536a28bf33e853b187dcae28894e869f43f7f5a9c3349a4c1  ./kata-ai-triage-brain-dump/input.schema.json
+527b5c1c35e8c82b3b656fc037b69721df3e5576a661d6fd03f154efd75269e3  ./kata-ai-triage-brain-dump/output.schema.json
+25290f5663dd71a212a1c2295d9094f86db288f6f0e48aaca68465c6472604e7  ./kata-ai-unstick-breakdown/fixtures/01-dreaded-email.json
+657391ba781b2f802191f561fc285958e2e08a905daaec3b7ef3f2cc7a1aaec3  ./kata-ai-unstick-breakdown/fixtures/02-laundry.json
+18fd63cbb4c8ea514dbdd851c6824c9b2eee35413ab986cbc231f502b767d383  ./kata-ai-unstick-breakdown/input.schema.json
+6333b7a54bbec42465f4a6533b58fce87995f7ac767107cdd956b488bd571ac9  ./kata-ai-unstick-breakdown/output.schema.json
+b2f54371cb084a3687646e13a406919f8f180ad96e6558812c247cb2a8960465  ./kata-overwhelm/fixtures/01-thesis-stuck.json
+b655ce075478f02d3aa236345f12687d57fc0add5468c53021ffaf9266c2ba71  ./kata-overwhelm/fixtures/02-goal-only.json
+cd37cb1c8a6b49dfa36aedb84a97157837720b8fa34738fe2bedfa4158645f0c  ./kata-overwhelm/input.schema.json
+69289ea7cd089d1598c098c2e7865038ff579242a10b8f91a1d1231ea5f48f04  ./kata-overwhelm/output.schema.json
+8157cdb2c076d1b754fd0884f5f0cc06ffbf1cf9fc00d5388b5d16015b31b8af  ./kata-roll-seed/fixtures/01-stuck-on-email.json
+1ac3ac3ceeb82012f0de1feb751db5f4f4566be189fb42a7aa4de752a8f68653  ./kata-roll-seed/fixtures/02-nothing-stuck.json
+1a4435ae4b1c485510845ca8a62f66150255bc117067f76f7f53e8c75ef8f379  ./kata-roll-seed/input.schema.json
+5250fc5bba62d82409783449c6e2a1c8e637c73dea4dbee53953d7d2b6a8e9f6  ./kata-roll-seed/output.schema.json
+70e6634249bb8f234dc6ab548b5d83d067b94abfb9c18153ebab72e8be8cd988  ./lint.json
diff --git a/plugins/kata/contracts/kata-ai/v1/README.md b/plugins/kata/contracts/kata-ai/v1/README.md
@@ -0,0 +1,39 @@
+# kata-ai/v1
+
+The contract between Kata (the app) and the DojoGenesis gateway for AI-backed skills. Defined by DGS-149; the ruling is in `specs/rollgoal-parity/05-ai-gateway-contract-scout.md` section 5.
+
+13 skills: the 11 `kata-ai-*` ops plus the two existing skills `kata-overwhelm` and `kata-roll-seed`.
+
+## Layout
+
+```
+_defs.schema.json            shared fragments (enums, TaskDraft, Milestone); human source of truth
+lint.json                    shame-aware voice lint rules
+<skill-id>/input.schema.json
+<skill-id>/output.schema.json
+<skill-id>/fixtures/NN-<slug>.json
+```
+
+Fixture shape: `{ description, input, valid_output, invalid_outputs: [{ why, output }] }`. Each `why` starts with `schema:` (must fail the output schema) or `lint:` (must pass the schema but trip a lint rule). Every `valid_output` must pass its schema and trip no lint rule.
+
+## Rules
+
+- JSON Schema draft 2020-12. Outputs are always an object at the root, because the gateway uses the output schema verbatim as a function-tool `parameters` object.
+- Output schemas inline the shared defs (no `$ref`, no `oneOf`/`anyOf`) so they work as tool parameters across providers. `_defs.schema.json` is the human source; if it changes, update every output that inlines the fragment. Input schemas `$ref` it.
+- `additionalProperties: false` everywhere. Sizes are capped on purpose (steps 1-7, milestones 1-10, tasks 1-12, titles at 120 characters, steps at 200, intention at 160).
+- `user_id` appears only in `kata-overwhelm` and `kata-roll-seed`. The 11 `kata-ai-*` ops get no user_id and no memory context in v1.
+- Enums mirror `src/types/enums.ts` and `src/types/work.ts`. `priority` on a goal is the `GoalPriority` string enum, not a number.
+
+## Gateway vendoring
+
+The gateway keeps a copy of this directory and runs a CI check that the file hashes match Kata's. It validates every model response against the output schema, and the same fixtures drive Kata's parser contract tests.
+
+## Lint rules
+
+`lint.json` applies to every string value in an output, recursively. Patterns are RE2 compatible (Go `regexp`): no lookarounds. Rules:
+
+no-exclamation, no-dont-worry, no-youve-got-this, no-you-should, no-you-didnt, no-you-failed, no-lazy, no-just-do-it, no-simply, no-medical-diagnos, no-medical-treat, no-medical-cure, no-medical-symptom.
+
+## Versioning
+
+A breaking change (removed or renamed key, narrowed enum, tighter bound that rejects previously valid output) goes in a new `v2/` directory. v1 stays frozen. Additive, backward-compatible loosening may land in v1.
diff --git a/plugins/kata/contracts/kata-ai/v1/SOURCE b/plugins/kata/contracts/kata-ai/v1/SOURCE
@@ -0,0 +1,5 @@
+source: DojoGenesis/kata contracts/kata-ai/v1
+commit: bde2add7be8d4adbb309db1f25cf87dd8f60c445
+branch: feat/dgs-149-kata-ai-v1-contracts
+vendored: 2026-10-02
+rule: do not edit here; change in Kata, re-vendor, regenerate MANIFEST.sha256 (shasum -a 256 over every file except MANIFEST.sha256 and SOURCE)
diff --git a/plugins/kata/contracts/kata-ai/v1/_defs.schema.json b/plugins/kata/contracts/kata-ai/v1/_defs.schema.json
@@ -0,0 +1,184 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://contracts.dojogenesis.com/kata-ai/v1/_defs.schema.json",
+  "title": "kata-ai/v1 shared definitions",
+  "description": "Human source of truth for shared fragments. Mirrors src/types/enums.ts and src/types/work.ts. Output schemas inline these (no $ref) so they work as function-tool parameters; input schemas $ref them.",
+  "$defs": {
+    "EnergyLevel": {
+      "type": "string",
+      "enum": [
+        "low",
+        "medium",
+        "high"
+      ]
+    },
+    "Urgency": {
+      "type": "string",
+      "enum": [
+        "low",
+        "medium",
+        "high",
+        "must_do"
+      ]
+    },
+    "Vibe": {
+      "type": "string",
+      "enum": [
+        "calm",
+        "creative",
+        "admin",
+        "deep_work",
+        "maintenance",
+        "social"
+      ]
+    },
+    "TaskSize": {
+      "type": "string",
+      "enum": [
+        "tiny",
+        "small",
+        "medium",
+        "large"
+      ]
+    },
+    "GoalPriority": {
+      "type": "string",
+      "enum": [
+        "low",
+        "medium",
+        "high",
+        "critical"
+      ]
+    },
+    "RoutineType": {
+      "type": "string",
+      "enum": [
+        "morning",
+        "evening",
+        "workout",
+        "chores",
+        "custom"
+      ]
+    },
+    "CelebrationType": {
+      "type": "string",
+      "enum": [
+        "restorative",
+        "energizing",
+        "social",
+        "sensory"
+      ]
+    },
+    "BrainDumpCategory": {
+      "type": "string",
+      "enum": [
+        "task",
+        "goal",
+        "routine",
+        "celebration",
+        "milestone",
+        "mixed"
+      ]
+    },
+    "TaskStatus": {
+      "type": "string",
+      "enum": [
+        "active",
+        "completed",
+        "archived",
+        "blocked"
+      ]
+    },
+    "TaskDraft": {
+      "type": "object",
+      "description": "A small, editable task suggestion. Only title is required.",
+      "properties": {
+        "title": {
+          "type": "string",
+          "maxLength": 120,
+          "minLength": 1
+        },
+        "description": {
+          "type": "string",
+          "maxLength": 300
+        },
+        "time_needed": {
+          "type": "number",
+          "exclusiveMinimum": 0,
+          "maximum": 480,
+          "description": "Minutes."
+        },
+        "energy_required": {
+          "type": "string",
+          "enum": [
+            "low",
+            "medium",
+            "high"
+          ]
+        },
+        "urgency": {
+          "type": "string",
+          "enum": [
+            "low",
+            "medium",
+            "high",
+            "must_do"
+          ]
+        },
+        "vibe": {
+          "type": "string",
+          "enum": [
+            "calm",
+            "creative",
+            "admin",
+            "deep_work",
+            "maintenance",
+            "social"
+          ]
+        },
+        "size": {
+          "type": "string",
+          "enum": [
+            "tiny",
+            "small",
+            "medium",
+            "large"
+          ]
+        },
+        "tags": {
+          "type": "array",
+          "items": {
+            "type": "string",
+            "maxLength": 40,
+            "minLength": 1
+          },
+          "minItems": 0,
+          "maxItems": 5
+        }
+      },
+      "required": [
+        "title"
+      ],
+      "additionalProperties": false
+    },
+    "Milestone": {
+      "type": "object",
+      "description": "Draft milestone. The app adds completed:false when it persists.",
+      "properties": {
+        "title": {
+          "type": "string",
+          "maxLength": 120,
+          "minLength": 1
+        },
+        "description": {
+          "type": "string",
+          "maxLength": 300
+        }
+      },
+      "required": [
+        "title"
+      ],
+      "additionalProperties": false
+    }
+  }
+}