eai-cli 0.0.0-bootstrap.0

package/resources/gofer/copilot-prompts/gofer_bootstrap_workspace.prompt.md

@@ -0,0 +1,161 @@
+---
+name: gofer:bootstrap-workspace
+description: Create or update the repo-owned Gofer scaffold for the current workspace.
+agent: copilot-workspace
+tools:
+  - Read
+  - Grep
+  - Glob
+  - Bash
+  - WebSearch
+argument-hint: feature-name-or-description
+gofer:
+  workflowProfile: standard
+  canonicalSource: .specify/commands/gofer_bootstrap_workspace.md
+  canonicalChecksum: 49ba02c80cf374af77f40e10fee49b4c71c389a2da01b402480b2e5e7b607b03
+  metadataSource: scripts/generate-commands.ts
+---
+
+
+# Gofer Workspace Bootstrap
+
+## Token And Cost Policy
+<!-- gofer:token-cost-policy:start -->
+
+Before spawning agents, calling tools, or loading large files:
+
+1. Treat `.specify/memory/gofer-model-policy.yaml` as the repo-owned source of truth for simple, medium, hard, and arbiter model routing. If it is missing, run `/gofer:bootstrap-workspace` before continuing.
+2. Use the cheapest capable model first.
+   - Claude: Haiku for scouting/extraction; Sonnet for normal implementation, synthesis, validation, and security; Opus for high-risk arbitration or release-critical failures.
+   - Codex/OpenAI: GPT mini for simple coding; GPT nano only for locate/classify/summarize/mechanical work; GPT-5.3-Codex or flagship GPT for tool-heavy coding, architecture, and release-critical validation.
+   - Gemini: Flash-Lite for cheap large-context scan/summarize; Flash for default research synthesis; Pro for large-context architecture or high-risk arbitration.
+   - Copilot: prefer Auto for simple and default work; ask the user before choosing a paid/high-tier picker model for hard security, architecture, or release gates.
+3. Keep raw tool output out of the main conversation context. Save stable findings to `.specify/specs/{feature}/context-bundle.md`, then work from summaries.
+4. Use provider prompt/context caching only for stable, non-secret prefixes: Gofer scaffold, AGENTS/CLAUDE/Copilot instructions, constitution, repo map, stage contracts, and validation rubric.
+5. Before continuing after large research, planning, implementation, or validation bursts, checkpoint the durable artifacts and compact/clear/resume context when the host supports it.
+6. Escalate model tier only when a cheaper pass is low-confidence, contradictory, security-sensitive, or blocking release quality.
+<!-- gofer:token-cost-policy:end -->
+
+## Guardrails
+
+- Preserve existing `.specify/specs/*` work.
+- Preserve existing `.specify/memory/constitution.md` if present.
+- Do **not** overwrite existing `AGENTS.md`, `CLAUDE.md`, or
+  `.github/copilot-instructions.md` by default. Create them only when missing.
+- Do **not** add repo-local assistant mirrors (`.claude/commands`,
+  `.agents/skills`, `.github/prompts`, `.gemini/`, etc.) unless the user
+  explicitly asks for them.
+
+## Step 1: Resolve The Workspace Root
+
+Use the current repository root. If you started inside a subdirectory, walk
+upward to the nearest directory containing `.git`, `package.json`,
+`pyproject.toml`, `go.mod`, `Cargo.toml`, or `.specify`.
+
+## Step 2: Select A Bootstrap Source
+
+Use the first source that exists:
+
+1. `~/plugins/eai-gofer`
+2. `./plugins/eai-gofer` when you are working in the Gofer repo itself
+3. Download the latest public bundle zip and extract it to a temporary folder:
+
+```bash
+curl -fsSL https://eai-tools.github.io/eai-gofer/releases/eai-gofer-agent-plugin-latest.zip \
+  -o /tmp/eai-gofer-agent-plugin-latest.zip
+rm -rf /tmp/eai-gofer-bootstrap
+mkdir -p /tmp/eai-gofer-bootstrap
+unzip -q /tmp/eai-gofer-agent-plugin-latest.zip -d /tmp/eai-gofer-bootstrap
+```
+
+When using the downloaded bundle, the source root is:
+
+```text
+/tmp/eai-gofer-bootstrap/eai-gofer
+```
+
+## Step 3: Run The Portable Bootstrap Script
+
+Prefer the scripted bootstrap from the selected source:
+
+```bash
+node "$SOURCE_ROOT/.specify/scripts/node/gofer-workspace-bootstrap.mjs" \
+  --workspace "$WORKSPACE_ROOT" \
+  --host "$GOFER_HOST"
+```
+
+Use the host value that matches the current client: `claude`, `codex`,
+`copilot`, or `gemini`.
+
+If the user explicitly asks for repo-local assistant mirrors as well, append:
+
+```bash
+--include-mirrors
+```
+
+The default bootstrap must create/update:
+
+- `.specify/commands/`
+- `.specify/templates/`
+- `.specify/scripts/bash/`
+- `.specify/scripts/node/`
+- `.specify/scripts/hooks/`
+- `.specify/scripts/powershell/`
+- `.specify/specs/`
+- `.specify/memory/`
+- `.specify/memory/gofer-model-policy.yaml` when missing
+- `.specify/.gofer-version`
+- `.specify/README.md`
+- Gofer runtime `.gitignore` entries
+
+And host-specific repo-owned files:
+
+- **Claude**: `AGENTS.md`, `CLAUDE.md`, `.claude/settings.json`
+- **Codex**: `AGENTS.md`
+- **Copilot**: `.github/copilot-instructions.md`
+- **Gemini**: no extra required repo-owned file beyond the core scaffold
+
+## Step 4: Re-Run The Workspace Check
+
+After bootstrap completes, run:
+
+```bash
+node .specify/scripts/node/gofer-workspace-check.mjs --host "$GOFER_HOST" --json
+```
+
+Report what changed and whether the workspace is now healthy.
+
+If the workspace is still stale because the repo-local scripts are older than
+the currently installed plugin bundle, explain that clearly and keep the
+selected source bundle as the truth for the update.
+
+## Step 5: Continue EAI App First-Run Setup When Needed
+
+If the user is building an EAI Platform app and Git, Node.js, npm, the EAI CLI,
+login, tenant access, or EAI app template readiness is still missing, run
+`/gofer:eai-first-run` next. Do not ask again when the workspace is healthy;
+only ask when an install, browser login, tenant selection, or `eai init` action
+is actually needed.
+
+## Step 6: Write The Workspace Bootstrap Report
+
+Write the artifact only to `.specify/logs/workspace-bootstrap-report.md`.
+
+If the target file already exists, replace it and prepend a regeneration note
+such as `<!-- regenerated at [ISO timestamp] -->`.
+
+Include the minimum provenance schema:
+
+- `GeneratedAt`
+- `SourceCommandId`
+- `SourceInputs`
+- `OverwriteNoticeWhenApplicable`
+
+The generated workspace bootstrap report must contain these sections:
+
+- `## Provenance`
+- `## Workspace Root`
+- `## Bootstrap Source`
+- `## Host Policy`
+- `## Changes Applied`
+- `## Post-Check`

package/resources/gofer/copilot-prompts/gofer_check_workspace.prompt.md

@@ -0,0 +1,132 @@
+---
+name: gofer:check-workspace
+description: Check whether this repo is initialized for Gofer and explain any missing or stale scaffold.
+agent: copilot-workspace
+tools:
+  - Read
+  - Grep
+  - Glob
+  - Bash
+  - WebSearch
+argument-hint: feature-name-or-description
+gofer:
+  workflowProfile: standard
+  canonicalSource: .specify/commands/gofer_check_workspace.md
+  canonicalChecksum: 95e354887cb2afc3c2e17c90bd10db8f166ce863c7543cde8f635d40058022f7
+  metadataSource: scripts/generate-commands.ts
+---
+
+
+# Gofer Workspace Check
+
+## Token And Cost Policy
+<!-- gofer:token-cost-policy:start -->
+
+Before spawning agents, calling tools, or loading large files:
+
+1. Treat `.specify/memory/gofer-model-policy.yaml` as the repo-owned source of truth for simple, medium, hard, and arbiter model routing. If it is missing, run `/gofer:bootstrap-workspace` before continuing.
+2. Use the cheapest capable model first.
+   - Claude: Haiku for scouting/extraction; Sonnet for normal implementation, synthesis, validation, and security; Opus for high-risk arbitration or release-critical failures.
+   - Codex/OpenAI: GPT mini for simple coding; GPT nano only for locate/classify/summarize/mechanical work; GPT-5.3-Codex or flagship GPT for tool-heavy coding, architecture, and release-critical validation.
+   - Gemini: Flash-Lite for cheap large-context scan/summarize; Flash for default research synthesis; Pro for large-context architecture or high-risk arbitration.
+   - Copilot: prefer Auto for simple and default work; ask the user before choosing a paid/high-tier picker model for hard security, architecture, or release gates.
+3. Keep raw tool output out of the main conversation context. Save stable findings to `.specify/specs/{feature}/context-bundle.md`, then work from summaries.
+4. Use provider prompt/context caching only for stable, non-secret prefixes: Gofer scaffold, AGENTS/CLAUDE/Copilot instructions, constitution, repo map, stage contracts, and validation rubric.
+5. Before continuing after large research, planning, implementation, or validation bursts, checkpoint the durable artifacts and compact/clear/resume context when the host supports it.
+6. Escalate model tier only when a cheaper pass is low-confidence, contradictory, security-sensitive, or blocking release quality.
+<!-- gofer:token-cost-policy:end -->
+
+## Step 1: Resolve The Workspace Root
+
+Use the current working directory unless you are already inside `.specify/` or a
+subdirectory. If needed, walk upward to the nearest directory containing one of:
+
+- `.git`
+- `package.json`
+- `pyproject.toml`
+- `go.mod`
+- `Cargo.toml`
+- `.specify`
+
+## Step 2: Check The Core Gofer Sentinels
+
+Inspect these paths relative to the workspace root:
+
+- `.specify/.gofer-version`
+- `.specify/commands#0_gofer_start.md`
+- `.specify/templates/spec-template.md`
+- `.specify/scripts/bash/create-new-feature.sh`
+- `.specify/scripts/node/parse-stage-command.mjs`
+- `.specify/scripts/hooks/post-tool-use.mjs`
+- `.specify/scripts/powershell/install-optional-tools.ps1`
+- `.specify/templates/gofer-model-policy.yaml`
+- `.specify/memory/gofer-model-policy.yaml`
+- `.specify/specs/`
+- `.specify/memory/`
+- `.specify/memory/gofer-model-policy.yaml` (create with `/gofer:bootstrap-workspace` if missing)
+
+## Step 3: Check Host-Specific Files
+
+Check the current host's required repo-owned files:
+
+- **Claude**: `AGENTS.md`, `CLAUDE.md`, `.claude/settings.json`
+- **Codex**: `AGENTS.md`
+- **Copilot**: `.github/copilot-instructions.md`
+- **Gemini**: no additional required repo-owned files beyond the core scaffold
+
+## Step 4: Prefer Scripted Evidence When Available
+
+If the repo already has the Gofer workspace scripts, run the checker for the
+current host and report the JSON result. Use the host value that matches the
+client you are currently in: `claude`, `codex`, `copilot`, or `gemini`.
+
+```bash
+node .specify/scripts/node/gofer-workspace-check.mjs --host "$GOFER_HOST" --json
+```
+
+If that script is missing, perform the equivalent manual path checks yourself
+and summarize the result in the same categories:
+
+- `healthy`
+- `missing`
+- `stale`
+
+Treat the workspace as `stale` when `.specify/.gofer-version` is present but
+does not match the installed Gofer/plugin version you are currently running.
+
+## Step 5: Write The Workspace Check Report
+
+Write the artifact only to `.specify/logs/workspace-check-report.md`.
+
+If the target file already exists, replace it and prepend a regeneration note
+such as `<!-- regenerated at [ISO timestamp] -->`.
+
+Include the minimum provenance schema:
+
+- `GeneratedAt`
+- `SourceCommandId`
+- `SourceInputs`
+- `OverwriteNoticeWhenApplicable`
+
+The generated workspace check report must contain these sections:
+
+- `## Provenance`
+- `## Workspace Root`
+- `## Core Scaffold`
+- `## Host Requirements`
+- `## Status`
+- `## Recommendation`
+
+## Step 6: Report And Ask Once If Repair Is Needed
+
+If the workspace is healthy, say so briefly and continue.
+
+If the workspace is missing or stale, ask exactly:
+
+**"This repo is missing or stale for Gofer. Initialize/update it now?"**
+
+If the user says **yes**, run `/gofer:bootstrap-workspace` next.
+
+If the user says **no**, stop. Explain that Gofer stage/helper commands depend
+on the repo-owned scaffold and should not continue until the repo is initialized
+or updated.