gsd-pi 2.76.0-dev.fe143342a → 2.77.0-dev.1d17f366c

package/src/resources/skills/create-skill/SKILL.md

@@ -88,8 +88,8 @@ Then proceed directly to the workflow.
 ## Skill Structure Quick Reference
 
 **Skill directories:**
-- Global: `~/.gsd/agent/skills/{skill-name}/`
-- Project-local: `.pi/agent/skills/{skill-name}/`
+- Global: `~/.agents/skills/{skill-name}/`
+- Project-local: `.agents/skills/{skill-name}/`
 
 **Simple skill (single file):**
 ```yaml

package/src/resources/skills/create-skill/references/gsd-skill-ecosystem.md

@@ -5,13 +5,13 @@ GSD-specific skill ecosystem details: directory conventions, discovery mechanics
 <skill_directories>
 GSD supports two skill directories, checked in order:
 
-**User-scope (global):** `~/.gsd/agent/skills/`
+**User-scope (global):** `~/.agents/skills/`
 - Available in every GSD session regardless of working directory
-- Installed by default or by the user
+- Installed via [skills.sh](https://skills.sh) or manually
 
-**Project-scope (local):** `.pi/agent/skills/`
+**Project-scope (local):** `.agents/skills/`
 - Available only when GSD runs inside the project directory
-- The project-local directory uses `.pi` (inherited from the pi base), **not** `.gsd`
+- Committable to version control so team members share the same skill set
 - Ideal for project-specific workflows, deploy scripts, or conventions
 
 Skills in both directories follow the same SKILL.md format and router pattern conventions.

package/src/resources/skills/create-skill/workflows/audit-skill.md

@@ -16,22 +16,22 @@
 Enumerate skills from both directories:
 ```bash
 echo "=== Global skills ==="
-ls ~/.gsd/agent/skills/ 2>/dev/null || echo "(none)"
+ls ~/.agents/skills/ 2>/dev/null || echo "(none)"
 
 echo "=== Project-local skills ==="
-ls .pi/agent/skills/ 2>/dev/null || echo "(none)"
+ls .agents/skills/ 2>/dev/null || echo "(none)"
 ```
 
 Present as:
 ```
 Available skills:
 
-Global (~/.gsd/agent/skills/):
+Global (~/.agents/skills/):
 1. create-skill
 2. manage-stripe
 ...
 
-Project-local (.pi/agent/skills/):
+Project-local (.agents/skills/):
 3. project-deploy
 ...
 ```

package/src/resources/skills/create-skill/workflows/create-new-skill.md

@@ -15,8 +15,8 @@
 **Ask the user:**
 "Should this skill be global or project-local?"
 
-1. **Global** (`~/.gsd/agent/skills/`) — Available in all GSD sessions
-2. **Project-local** (`.pi/agent/skills/`) — Available only in this project
+1. **Global** (`~/.agents/skills/`) — Available in all GSD sessions
+2. **Project-local** (`.agents/skills/`) — Available only in this project
 
 ## Step 2: Adaptive Requirements Gathering
 
@@ -99,11 +99,11 @@ See references/recommended-structure.md for templates.
 ## Step 5: Create Directory
 
 Use the scope selected in Step 1 to determine the base path:
-- **Global:** `~/.gsd/agent/skills/{skill-name}`
-- **Project-local:** `.pi/agent/skills/{skill-name}`
+- **Global:** `~/.agents/skills/{skill-name}`
+- **Project-local:** `.agents/skills/{skill-name}`
 
 ```bash
-BASE_PATH="{selected-base}/agent/skills/{skill-name}"
+BASE_PATH="{selected-base}/skills/{skill-name}"
 mkdir -p $BASE_PATH
 # If complex:
 mkdir -p $BASE_PATH/workflows

package/src/resources/skills/verify-before-complete/SKILL.md

@@ -1,6 +1,7 @@
 ---
 name: verify-before-complete
-description: Block completion claims until verification evidence has been produced in the current message. Use before marking a task/slice/milestone complete, before creating a commit or PR, before saying "it works" or "tests pass", and any time you are about to claim work is done. The rule is: evidence before claims, always — running the verification must happen now, not "earlier in the session". Fresh output or no claim.
+description: >-
+  Block completion claims until verification evidence has been produced in the current message. Use before marking a task/slice/milestone complete, before creating a commit or PR, before saying "it works" or "tests pass", and any time you are about to claim work is done. The rule is: evidence before claims, always — running the verification must happen now, not "earlier in the session". Fresh output or no claim.
 ---
 
 <objective>

package/src/resources/skills/write-docs/SKILL.md

@@ -1,6 +1,7 @@
 ---
 name: write-docs
-description: Collaborative document authoring workflow for proposals, technical specs, decision docs, README sections, ADRs, and long-form prose that must work for fresh readers. Use when asked to "write the docs", "draft a proposal", "write a spec", "write an RFC", "write the README", or when a document needs to be understandable by someone without this session's context. Three stages: gather context, iterate on structure, reader-test for a stranger.
+description: >-
+  Collaborative document authoring workflow for proposals, technical specs, decision docs, README sections, ADRs, and long-form prose that must work for fresh readers. Use when asked to "write the docs", "draft a proposal", "write a spec", "write an RFC", "write the README", or when a document needs to be understandable by someone without this session's context. Three stages: gather context, iterate on structure, reader-test for a stranger.
 ---
 
 <objective>

package/dist/web/standalone/.next/server/chunks/7461.js

@@ -1 +0,0 @@
-"use strict";exports.id=7461,exports.ids=[7461],exports.modules={7461:(a,b,c)=>{c.d(b,{isOnboardingComplete:()=>k,readOnboardingRecord:()=>j});var d=c(3024),e=c(8161),f=c(6760);c(7598),c(1421),process.env.GSD_ENABLE_NATIVE_GSD_PARSER,Symbol("native-unavailable"),c(9134),Object.values({Q3:{id:"Q3",scope:"slice",ownerTurn:"gate-evaluate",question:"How can this be exploited?",guidance:"Identify abuse scenarios: parameter tampering, replay attacks, privilege escalation.\nMap data exposure risks: PII, tokens, secrets accessible through this slice.\nDefine input trust boundaries: untrusted user input reaching DB, API, or filesystem.\nIf none apply, return verdict 'omitted' with rationale explaining why.",promptSection:"Abuse Surface"},Q4:{id:"Q4",scope:"slice",ownerTurn:"gate-evaluate",question:"What existing promises does this break?",guidance:"List which existing requirements (R001, R003, etc.) are touched by this slice.\nIdentify what must be re-tested after shipping.\nFlag decisions that should be revisited given the new scope.\nIf no existing requirements are affected, return verdict 'omitted'.",promptSection:"Broken Promises"},Q5:{id:"Q5",scope:"task",ownerTurn:"execute-task",question:"What breaks when dependencies fail?",guidance:"Enumerate the task's external dependencies (APIs, filesystem, network, subprocesses).\nDescribe the failure path for each: timeout, malformed response, connection loss.\nVerify the implementation handles each failure or explicitly bubbles the error.\nReturn verdict 'omitted' only if the task has no external dependencies.",promptSection:"Failure Modes"},Q6:{id:"Q6",scope:"task",ownerTurn:"execute-task",question:"What is the 10x load breakpoint?",guidance:"Identify the resource that saturates first at 10x the expected load.\nDescribe the protection applied (pool sizing, rate limiting, pagination, caching).\nReturn verdict 'omitted' if the task has no runtime load dimension.",promptSection:"Load Profile"},Q7:{id:"Q7",scope:"task",ownerTurn:"execute-task",question:"What negative tests protect this task?",guidance:"List malformed inputs, error paths, and boundary conditions the tests cover.\nPoint to the specific test files or cases that assert each negative scenario.\nReturn verdict 'omitted' only if the task has no meaningful negative surface.",promptSection:"Negative Tests"},Q8:{id:"Q8",scope:"slice",ownerTurn:"complete-slice",question:"How will ops know this slice is healthy or broken?",guidance:"Describe the health signal (metric, log line, dashboard) that proves the slice works.\nDescribe the failure signal that triggers an alert or paging.\nDocument the recovery procedure and any monitoring gaps.\nReturn verdict 'omitted' only for slices with no runtime behavior at all.",promptSection:"Operational Readiness"},MV01:{id:"MV01",scope:"milestone",ownerTurn:"validate-milestone",question:"Is every success criterion in the milestone roadmap satisfied?",guidance:"Walk the success-criteria checklist from the milestone roadmap.\nFor each criterion, point to the slice / assessment / verification evidence that proves it.\nReturn verdict 'flag' if any criterion is unmet or unverifiable.",promptSection:"Success Criteria Checklist"},MV02:{id:"MV02",scope:"milestone",ownerTurn:"validate-milestone",question:"Does every slice have a SUMMARY.md and a passing assessment?",guidance:"Confirm every slice listed in the roadmap has a SUMMARY.md.\nConfirm each slice has an ASSESSMENT verdict of 'pass' (or justified 'omitted').\nFlag missing artifacts and slices with outstanding follow-ups or known limitations.",promptSection:"Slice Delivery Audit"},MV03:{id:"MV03",scope:"milestone",ownerTurn:"validate-milestone",question:"Do the slices integrate end-to-end?",guidance:"Trace at least one cross-slice flow proving the pieces compose.\nFlag gaps where two slices were built in isolation with no integration evidence.",promptSection:"Cross-Slice Integration"},MV04:{id:"MV04",scope:"milestone",ownerTurn:"validate-milestone",question:"Are all touched requirements covered and still coherent?",guidance:"For each requirement advanced, validated, surfaced, or invalidated across the milestone's slices, confirm the milestone-level evidence matches.\nFlag requirements that slices claim to advance but no artifact proves.",promptSection:"Requirement Coverage"}});let g=process.env.GSD_CODING_AGENT_DIR||(0,f.join)(process.env.GSD_HOME||(0,f.join)((0,e.homedir)(),".gsd"),"agent"),h=(0,f.join)(g,"onboarding.json"),i={version:1,flowVersion:1,completedAt:null,completedSteps:[],skippedSteps:[],lastResumePoint:null};function j(){if(!(0,d.existsSync)(h))return{...i};try{let a=JSON.parse((0,d.readFileSync)(h,"utf-8"));return{version:"number"==typeof a.version?a.version:1,flowVersion:"number"==typeof a.flowVersion?a.flowVersion:0,completedAt:"string"==typeof a.completedAt?a.completedAt:null,completedSteps:Array.isArray(a.completedSteps)?a.completedSteps.filter(a=>"string"==typeof a):[],skippedSteps:Array.isArray(a.skippedSteps)?a.skippedSteps.filter(a=>"string"==typeof a):[],lastResumePoint:"string"==typeof a.lastResumePoint?a.lastResumePoint:null}}catch{return{...i}}}function k(){let a=j();return null!==a.completedAt&&1===a.flowVersion}},9134:(a,b,c)=>{c.d(b,{DZ:()=>e,rs:()=>d});let d="GSD_GIT_ERROR";class e extends Error{constructor(a,b,c){super(b,c),this.name="GSDError",this.code=a}}}};