repo-harness 0.7.0 → 0.7.1

package/README.es.md

@@ -1,7 +1,7 @@
 # repo-harness
 
 <p align="center">
-  <img src="docs/images/image.png" alt="One next button joining Claude and Codex under repo-harness workflow rules" width="760">
+  <img src="docs/images/repo-harness-gptpro.png" alt="repo-harness architecture and ChatGPT Pro local planner workflow diagram" width="960">
 </p>
 
 `repo-harness` convierte las sesiones de programación con Claude/Codex en un
@@ -56,7 +56,7 @@ En un repositorio adoptado, la superficie se mantiene pequeña:
 | `tasks/contracts/`, `tasks/reviews/` y `.ai/harness/checks/` | Scope, verificación y evidencia de review para probar que el trabajo terminó. |
 | `.ai/harness/handoff/` y `tasks/current.md` | Session journal y estado resumible, derivados de workflow artifacts en vez de chat memory. |
 
-## Novedades en 0.7.0
+## Novedades en 0.7.1
 
 - **ChatGPT browser engine.** `repo-harness chatgpt browser-*` crea sesiones
   ChatGPT Web repo-locales con archivos revisados por policy, sin usar OpenAI API.
@@ -383,8 +383,8 @@ Guards habituales:
 
 ## Release actual
 
-- npm package: `repo-harness@0.7.0`
-- Generated workflow stamp: `repo-harness@0.7.0+template@0.7.0`
+- npm package: `repo-harness@0.7.1`
+- Generated workflow stamp: `repo-harness@0.7.1+template@0.7.1`
 - GitHub repository: `Ancienttwo/repo-harness`
 - Release history: [`docs/CHANGELOG.md`](docs/CHANGELOG.md)
 

package/README.fr.md

@@ -1,7 +1,7 @@
 # repo-harness
 
 <p align="center">
-  <img src="docs/images/image.png" alt="One next button joining Claude and Codex under repo-harness workflow rules" width="760">
+  <img src="docs/images/repo-harness-gptpro.png" alt="repo-harness architecture and ChatGPT Pro local planner workflow diagram" width="960">
 </p>
 
 `repo-harness` transforme les sessions de code Claude/Codex en workflow
@@ -56,7 +56,7 @@ Dans un dépôt adopté, la surface à comprendre reste volontairement réduite
 | `tasks/contracts/`, `tasks/reviews/` et `.ai/harness/checks/` | Scope, vérification et preuves de review pour démontrer que le travail est terminé. |
 | `.ai/harness/handoff/` et `tasks/current.md` | Session journal et état resumable dérivés des workflow artifacts plutôt que de la chat memory. |
 
-## Nouveautés de la 0.7.0
+## Nouveautés de la 0.7.1
 
 - **ChatGPT browser engine.** `repo-harness chatgpt browser-*` crée des sessions
   ChatGPT Web repo-locales avec des fichiers contrôlés par policy, sans OpenAI API.
@@ -387,8 +387,8 @@ Guards courants :
 
 ## Release actuelle
 
-- npm package : `repo-harness@0.7.0`
-- Generated workflow stamp : `repo-harness@0.7.0+template@0.7.0`
+- npm package : `repo-harness@0.7.1`
+- Generated workflow stamp : `repo-harness@0.7.1+template@0.7.1`
 - GitHub repository : `Ancienttwo/repo-harness`
 - Release history : [`docs/CHANGELOG.md`](docs/CHANGELOG.md)
 

package/README.ja.md

@@ -1,7 +1,7 @@
 # repo-harness
 
 <p align="center">
-  <img src="docs/images/image.png" alt="One next button joining Claude and Codex under repo-harness workflow rules" width="760">
+  <img src="docs/images/repo-harness-gptpro.png" alt="repo-harness architecture and ChatGPT Pro local planner workflow diagram" width="960">
 </p>
 
 `repo-harness` は、Claude/Codex のコーディング session を、繰り返し使える
@@ -47,7 +47,7 @@ Agent に完全な PRD または Sprint を渡せば、あとは review and `nex
 | `tasks/contracts/`, `tasks/reviews/`, `.ai/harness/checks/` | 作業完了を証明する scope、verification、review evidence。 |
 | `.ai/harness/handoff/` と `tasks/current.md` | chat memory ではなく workflow artifacts から派生する session journal と resumable status。 |
 
-## 0.7.0 の新機能
+## 0.7.1 の新機能
 
 - **ChatGPT browser engine。** `repo-harness chatgpt browser-*` は、OpenAI API を使わず、policy-checked な repo-local ChatGPT Web consult session を作れます。
 - **Opt-in MCP browser tools。** `repo-harness mcp serve --enable-chatgpt-browser` のときだけ consult/read/list/continue/open tools を公開し、デフォルトでは無効です。
@@ -352,8 +352,8 @@ hook がブロックしたときは、まず terminal の構造化された出
 
 ## 現在の Release
 
-- npm package：`repo-harness@0.7.0`
-- Generated workflow stamp：`repo-harness@0.7.0+template@0.7.0`
+- npm package：`repo-harness@0.7.1`
+- Generated workflow stamp：`repo-harness@0.7.1+template@0.7.1`
 - GitHub repository：`Ancienttwo/repo-harness`
 - Release history：[`docs/CHANGELOG.md`](docs/CHANGELOG.md)
 

package/README.md

@@ -1,7 +1,7 @@
 # repo-harness
 
 <p align="center">
-  <img src="docs/images/image.png" alt="One next button joining Claude and Codex under repo-harness workflow rules" width="760">
+  <img src="docs/images/repo-harness-gptpro.png" alt="repo-harness architecture and ChatGPT Pro local planner workflow diagram" width="960">
 </p>
 
 `repo-harness` turns Claude/Codex coding sessions into a repeatable repo-local
@@ -75,18 +75,17 @@ Agents read source artifacts before derived summaries:
 `tasks/current.md` is only an orientation snapshot. If it disagrees with the
 active plan, contract, review, checks, or handoff, the source artifacts win.
 
-## What's New in 0.7.0
+## What's New in 0.7.1
 
-- **ChatGPT browser engine.** `repo-harness chatgpt browser-*` can create
-  policy-checked, repo-local ChatGPT Web consult sessions without using the
-  OpenAI API.
-- **Opt-in MCP browser tools.** `repo-harness mcp serve
-  --enable-chatgpt-browser` exposes consult/read/list/continue/open tools while
-  keeping them disabled by default.
-- **Oracle and native providers.** The engine supports an Oracle browser wrapper
-  plus a native installed-Chrome CDP spike for logged-in ChatGPT Web runs.
-- **Hosted CI and release smoke.** The repo now ships a GitHub CI gate and a
-  tarball install smoke that starts the packaged `repo-harness` binaries.
+- **GPT Pro setup path.** `repo-harness:gptpro_setup` guides
+  `gptpro_browser` browser/session setup and `gptpro_mcp` ChatGPT Connector MCP
+  setup while keeping ChatGPT Pro separate from API quota.
+- **GPT Pro consult path.** `repo-harness:gptpro` presents `gptpro
+  consult/read/continue/open` language over the existing ChatGPT Web browser
+  session engine.
+- **Trace-safe PostBash checks.** PostBash advisory metadata now writes to
+  `.ai/harness/checks/post-bash-latest.json`, leaving `latest.json` reserved for
+  authoritative `repo-harness-run-trace.v1` evidence.
 
 ## What repo-harness Does
 
@@ -176,10 +175,11 @@ The design has three layers:
    `.ai/harness/`, helper scripts, and `.ai/hooks/`.
 3. **Host adapters**: user-level `~/.claude/settings.json` and
    `~/.codex/hooks.json` route Claude/Codex events into `repo-harness-hook`.
-   The hook entrypoint exits silently for non-opt-in repos. For opted-in repos,
-   it resolves hooks central-first through the packaged install or
-   `~/.repo-harness/hooks/`, with repo policy able to pin self-host development
-   back to `.ai/hooks/*`.
+
+The hook entrypoint exits silently for non-opt-in repos. For opted-in repos,
+it resolves hooks central-first through the packaged install or
+`~/.repo-harness/hooks/`, with repo policy able to pin self-host development
+back to `.ai/hooks/*`.
 
 For `UserPromptSubmit`, the public adapter contract stays
 `repo-harness-hook UserPromptSubmit --route default`. The CLI route registry
@@ -550,8 +550,8 @@ Most common guards:
 
 ## Current Release
 
-- npm package: `repo-harness@0.7.0`
-- Generated workflow stamp: `repo-harness@0.7.0+template@0.7.0`
+- npm package: `repo-harness@0.7.1`
+- Generated workflow stamp: `repo-harness@0.7.1+template@0.7.1`
 - GitHub repository: `Ancienttwo/repo-harness`
 - Release history: [`docs/CHANGELOG.md`](docs/CHANGELOG.md)
 
@@ -638,6 +638,8 @@ skill discovery compatible while the CLI and hooks own execution:
 - Product planning layer: `repo-harness-prd` (activates `$geju`, then uses Claude-first `claude -p --model opus` drafting with Codex fallback to write upper-layer PRDs in `plans/prds/`)
 - Sprint program layer: `repo-harness-sprint` (ordered sprint backlogs in `plans/sprints/`, each row expanded with `$think` before the contract flow)
 - Goal session layer: `repo-harness-goal` / `repo-harness:goal` (prepares Codex/Claude `/goal` prompts from detailed PRD or Sprint artifacts and asks for those docs when missing)
+- GPT Pro local setup layer: `repo-harness-gptpro-setup` / `repo-harness:gptpro_setup` (guides `gptpro_browser` browser/session setup and `gptpro_mcp` ChatGPT Connector MCP setup while keeping ChatGPT Pro separate from API quota)
+- GPT Pro consult layer: `repo-harness-gptpro` / `repo-harness:gptpro` (uses `gptpro consult/read/continue/open` language while mapping to the local ChatGPT Web browser session engine)
 - Repo workflow actions: `repo-harness-ship`, `repo-harness-init`, `repo-harness-migrate`, `repo-harness-upgrade`, `repo-harness-capability`, `repo-harness-architecture`, `repo-harness-handoff`, `repo-harness-deploy`, `repo-harness-repair`, `repo-harness-check`
 - Branch project creation command: `repo-harness-scaffold`
 

package/README.zh-CN.md

@@ -1,7 +1,7 @@
 # repo-harness
 
 <p align="center">
-  <img src="docs/images/image.png" alt="One next button joining Claude and Codex under repo-harness workflow rules" width="760">
+  <img src="docs/images/repo-harness-gptpro-cn.png" alt="repo-harness x GPT Pro 本地 Planner Agent 推广图" width="960">
 </p>
 
 `repo-harness` 把 Claude/Codex 的 AI 编程会话变成可复用、可恢复、可检查的
@@ -67,7 +67,7 @@ Agent 先读 source artifacts，再读派生摘要：
 `tasks/current.md` 只是 orientation snapshot。如果它和 active plan、contract、
 review、checks 或 handoff 冲突，以 source artifacts 为准。
 
-## 0.7.0 新特性
+## 0.7.1 新特性
 
 - **ChatGPT browser engine。** `repo-harness chatgpt browser-*` 可以用
   policy-checked 文件输入创建 repo-local ChatGPT Web consult session，不走 OpenAI API。
@@ -442,8 +442,8 @@ hook block 工作时，先看 terminal 里的结构化输出。核心字段是
 
 ## 当前 Release
 
-- npm package：`repo-harness@0.7.0`
-- Generated workflow stamp：`repo-harness@0.7.0+template@0.7.0`
+- npm package：`repo-harness@0.7.1`
+- Generated workflow stamp：`repo-harness@0.7.1+template@0.7.1`
 - GitHub repository：`Ancienttwo/repo-harness`
 - Release history：[`docs/CHANGELOG.md`](docs/CHANGELOG.md)
 

package/SKILL.md

@@ -145,6 +145,8 @@ contract files:
 - `repo-harness-prd`: generate an upper-layer PRD in `plans/prds/`
 - `repo-harness-sprint`: plan a sprint backlog in `plans/sprints/` from a PRD or source spec, then expand each row with `$think` before the task-contract flow
 - `repo-harness-goal` / `repo-harness:goal`: prepare a Codex/Claude `/goal` prompt from detailed PRD or Sprint artifacts; request those documents before starting when missing
+- `repo-harness-gptpro-setup` / `repo-harness:gptpro_setup`: guide `gptpro_browser` ChatGPT Web browser/session setup and `gptpro_mcp` ChatGPT Connector MCP setup without treating GPT Pro as API quota
+- `repo-harness-gptpro` / `repo-harness:gptpro`: consult or continue GPT Pro through the local ChatGPT Web browser/session bridge while presenting `gptpro consult/read/continue/open` wording over the underlying `browser-*` engine commands
 
 Internal steps such as `hooks-init`, `docs-init`, and `create-project-dirs` are
 not public commands. They stay behind `init`, `scaffold`, `migrate`, and

package/assets/hooks/post-bash.sh

@@ -169,10 +169,7 @@ fi
 
 checks_file="$(workflow_checks_file)"
 post_bash_checks_file="$(dirname "$checks_file")/post-bash-latest.json"
-target_checks_file="$checks_file"
-if [[ -f "$checks_file" ]] && grep -Eq '"source"[[:space:]]*:[[:space:]]*"verify-sprint"' "$checks_file"; then
-  target_checks_file="$post_bash_checks_file"
-fi
+target_checks_file="$post_bash_checks_file"
 
 mkdir -p "$(dirname "$target_checks_file")"
 if [[ -n "$raw_output_path" ]]; then
@@ -205,10 +202,10 @@ cat > "$target_checks_file" <<EOF_CHECKS
 }
 EOF_CHECKS
 
-if [[ "$target_checks_file" != "$checks_file" ]]; then
+if [[ -f "$checks_file" ]]; then
   echo "[ChecksFile] Preserved ${checks_file}; updated ${target_checks_file}."
 else
-  echo "[ChecksFile] Updated ${checks_file}."
+  echo "[ChecksFile] Updated ${target_checks_file}; ${checks_file} remains reserved for repo-harness-run-trace.v1."
 fi
 
 # Aggregated advisory (route-registry keeps one PostToolUse bash entry; the

package/assets/reference-configs/agentic-development-flow.md

@@ -39,6 +39,8 @@ migrating, repairing, or verifying this repo-local harness:
 | Generate an upper-layer PRD | `repo-harness-prd` | `$geju` direction pass, Claude-first `claude -p --model opus` drafting, Codex fallback only when needed, PRD in `plans/prds/*.prd.md` |
 | Plan and run a program-level sprint | `repo-harness-sprint` | Upper-layer PRD in `plans/prds/`, sprint backlog in `plans/sprints/`; each row expands through `$think` before plan -> contract -> worktree |
 | Prepare a bounded native goal session | `repo-harness-goal` / `repo-harness:goal` | Codex/Claude `/goal` prompt from detailed PRD or Sprint artifacts; stops to request those documents when missing |
+| Configure GPT Pro local bridge | `repo-harness-gptpro-setup` / `repo-harness:gptpro_setup` | Separates `gptpro_browser` local ChatGPT Web browser/session consults from `gptpro_mcp` ChatGPT Connector MCP sidecar setup; preserves auth, tunnel, and API-billing boundaries |
+| Consult GPT Pro through browser session | `repo-harness-gptpro` / `repo-harness:gptpro` | Uses `gptpro consult/read/continue/open` wording while mapping to `browser-consult`, `browser-session`, `browser-followup`, and `browser-open` engine commands |
 
 `hooks-init`, `docs-init`, and `create-project-dirs` are not public commands.
 They are implementation steps behind `init`, `scaffold`, `migrate`, and

package/assets/skill-commands/manifest.json

@@ -122,6 +122,20 @@
       "mutatesRepoByDefault": false,
       "engine": "Codex or Claude native /goal plus repo-harness PRD/Sprint contract",
       "summary": "Prepares a bounded host-native /goal prompt from detailed PRD or Sprint artifacts and requires those documents before continuation."
+    },
+    {
+      "name": "repo-harness-gptpro-setup",
+      "class": "gptpro-local-setup",
+      "mutatesRepoByDefault": true,
+      "engine": "repo-harness chatgpt browser/session commands plus repo-harness MCP Connector setup",
+      "summary": "Guides gptpro_browser and gptpro_mcp setup so local repo-harness can use ChatGPT Web sessions and ChatGPT can connect back through MCP."
+    },
+    {
+      "name": "repo-harness-gptpro",
+      "class": "gptpro-browser-consult",
+      "mutatesRepoByDefault": true,
+      "engine": "repo-harness chatgpt browser-consult/session/followup/open using GPT Pro wording",
+      "summary": "Consults or continues GPT Pro through the local ChatGPT Web browser/session bridge while mapping user-facing gptpro language to browser engine commands."
     }
   ],
   "nonPublicInternalSteps": [

package/assets/skill-commands/repo-harness-gptpro/SKILL.md

@@ -0,0 +1,52 @@
+---
+name: repo-harness-gptpro
+description: Direct GPT Pro consult facade for repo-harness. Use when the user asks for repo-harness:gptpro, GPT Pro advice, GPT Pro review, gptpro consult, or continuing/opening a GPT Pro browser session through the local ChatGPT Web browser/session bridge.
+when_to_use: "repo-harness-gptpro, repo-harness:gptpro, gptpro, gptpro consult, GPT Pro consult, GPT Pro review, browser_session consult, browser-session consult, continue GPT Pro session"
+---
+
+# repo-harness-gptpro
+
+Use this command when the user wants local Codex to consult GPT Pro through the repo-harness ChatGPT Web browser/session bridge.
+
+Use GPT Pro language with the user. Treat the `browser-*` command names as implementation details:
+
+| User-facing action | Engine command |
+| --- | --- |
+| `gptpro consult` | `repo-harness chatgpt browser-consult` |
+| `gptpro continue` | `repo-harness chatgpt browser-followup` |
+| `gptpro read` | `repo-harness chatgpt browser-session` |
+| `gptpro open` | `repo-harness chatgpt browser-open` |
+| `gptpro list` | `repo-harness chatgpt browser-list` |
+
+## Protocol
+
+1. Confirm the target repo path with `git rev-parse --show-toplevel` or the user's explicit path. Preserve unrelated dirty worktree state.
+2. Verify GPT Pro browser readiness before any real consult:
+   `repo-harness chatgpt browser-doctor --repo <repo> --provider native --json`
+3. If setup is missing, route to `repo-harness-gptpro-setup` / `repo-harness:gptpro_setup` instead of improvising configuration.
+4. If the user provides an existing GPT Pro session id, inspect it with:
+   `repo-harness chatgpt browser-session --repo <repo> <sessionId>`
+5. Continue an existing GPT Pro conversation with:
+   `repo-harness chatgpt browser-followup --repo <repo> --session <sessionId> --prompt <prompt>`
+6. Open the saved GPT Pro conversation when the user asks to view it:
+   `repo-harness chatgpt browser-open --repo <repo> <sessionId>`
+7. Start a new GPT Pro consult with GPT Pro wording in the report, while running the browser engine underneath:
+   `repo-harness chatgpt browser-consult --repo <repo> --provider native --manual-login --prompt <prompt> --write-output .ai/harness/handoff/gptpro-consult.md`
+8. Use `--dry-run` first when attaching files, then run the real consult only after the prompt bundle and allowed paths are clear.
+9. Report the GPT Pro result with: session id, output path, conversation URL if present, whether it was a new consult or follow-up, and any manual-login blocker.
+
+## Failure Modes
+
+- If `browser-doctor` reports native provider unavailable, route to `repo-harness:gptpro_setup` and stop before claiming GPT Pro access works.
+- If ChatGPT Web needs login, captcha, SSO, workspace selection, or manual verification, report the blocker and ask the user to complete it in the visible browser.
+- If the user asks for `gptpro_mcp`, Connector setup, or ChatGPT -> local repo access, route to `repo-harness:gptpro_setup`; this command is for local -> GPT Pro consults.
+- If a session id is invalid or missing, list recent GPT Pro sessions with `repo-harness chatgpt browser-list --repo <repo>` and ask for the intended session.
+- If files include secrets, denied paths, symlink escapes, or oversized prompt bundles, preserve the failed dry-run evidence and do not run the real consult.
+
+## Boundaries
+
+- Does not rename or replace the underlying `repo-harness chatgpt browser-*` CLI commands; it only presents GPT Pro wording at the skill layer.
+- Does not configure MCP Connector, HTTPS tunnels, OAuth passphrases, or bearer tokens; use `repo-harness:gptpro_setup` for setup.
+- Does not request or handle ChatGPT passwords, 2FA codes, cookies, browser storage, or session tokens.
+- Does not use ChatGPT Pro as an OpenAI API key, API quota, or billing surface.
+- Does not treat GPT Pro advice as implementation authority; Codex still owns repo edits and verification.

package/assets/skill-commands/repo-harness-gptpro-setup/SKILL.md

@@ -0,0 +1,50 @@
+---
+name: repo-harness-gptpro-setup
+description: Guides GPT Pro local setup for repo-harness by configuring gptpro_browser ChatGPT Web browser/session consults and gptpro_mcp ChatGPT Connector MCP sidecar access with verification, auth, tunnel, and API-billing boundaries.
+when_to_use: "repo-harness-gptpro-setup, repo-harness:gptpro_setup, gptpro_setup, gptpro_browser, gptpro_broswser, gptpro_mcp, GPT Pro browser setup, ChatGPT Connector MCP setup"
+---
+
+# repo-harness-gptpro-setup
+
+Use this command when the user wants repo-harness to guide local GPT Pro setup across both directions:
+
+- `gptpro_browser`: local repo-harness calls an already logged-in ChatGPT Web session through browser/session automation.
+- `gptpro_mcp`: ChatGPT connects back to the local repo through the repo-harness MCP sidecar and Connector setup.
+
+## Protocol
+
+1. Confirm the target repo path with `git rev-parse --show-toplevel` or an explicit user path. Preserve unrelated dirty worktree state.
+2. State the two-lane model before configuring anything:
+   - `gptpro_browser` is local -> ChatGPT Web through `repo-harness chatgpt browser-*`.
+   - `gptpro_mcp` is ChatGPT -> local through `repo-harness mcp serve --transport http`.
+   - ChatGPT Pro Web access is not OpenAI API quota or an API key substitute.
+3. Configure browser/session support:
+   - `repo-harness chatgpt browser-setup --repo <repo>`
+   - `repo-harness chatgpt browser-doctor --repo <repo> --provider native --json`
+   - For a non-mutating prompt/file preview, run `repo-harness chatgpt browser-consult --repo <repo> --provider native --dry-run --prompt <text>`.
+4. For a real GPT Pro browser consult, require an already logged-in ChatGPT Web session and run a bounded command such as:
+   - `repo-harness chatgpt browser-consult --repo <repo> --provider native --manual-login --prompt <text> --model <label> --thinking <level>`
+5. Configure ChatGPT Connector MCP support:
+   - `repo-harness mcp setup chatgpt --repo <repo>`
+   - `repo-harness mcp doctor --repo <repo> --json`
+   - Start the sidecar when the user is ready to connect ChatGPT:
+     `repo-harness mcp serve --repo <repo> --transport http --host 127.0.0.1 --port 8765 --profile planner --enable-chatgpt-browser`
+6. Tell the user that ChatGPT Connector setup still requires an HTTPS tunnel or equivalent public HTTPS `/mcp` endpoint, then manual connector creation in ChatGPT settings. Keep OAuth passphrases and bearer tokens redacted.
+7. If local Codex also needs repo-harness MCP tools, separately run `repo-harness mcp setup codex --repo <repo> --scope project` and verify with `repo-harness mcp doctor --repo <repo> --json`.
+8. Finish with a concise matrix: `gptpro_browser` status, `gptpro_mcp` status, exact commands run, manual steps remaining, verification evidence, and residual risk.
+
+## Failure Modes
+
+- If `browser-doctor` reports the native provider unavailable, report the missing Chrome/CDP prerequisite and stop before a non-dry-run browser consult.
+- If ChatGPT Web is not logged in or requires manual verification, report the manual-login blocker and preserve the dry-run session artifact.
+- If `mcp doctor` reports `ready_local` but ChatGPT is not connected, report the missing HTTPS tunnel or manual Connector step instead of claiming end-to-end success.
+- If the user asks to use a ChatGPT Pro subscription as an OpenAI API key or API billing source, stop and explain that ChatGPT Web subscriptions and API Platform usage are separate products.
+- If any command would print `.repo-harness/mcp.tokens.json`, `.repo-harness/mcp.oauth.json`, browser profile secrets, or cookies, redact the value and report only the file class.
+
+## Boundaries
+
+- Does not create OpenAI API keys, API billing projects, or API credentials from a ChatGPT Pro subscription.
+- Does not bypass ChatGPT Web rate limits, login checks, manual verification, or plan restrictions.
+- Does not expose a local MCP server to the public internet without explicit auth, tunnel, and user intent.
+- Does not enable `--enable-chatgpt-browser` silently; require the user to ask for GPT Pro browser/session bridging.
+- Does not commit `.repo-harness/` local auth files, browser profiles, cookies, tokens, or tunnel state.

package/assets/skill-version.json

@@ -1,6 +1,6 @@
 {
-  "version": "0.7.0",
-  "templateVersion": "0.7.0",
+  "version": "0.7.1",
+  "templateVersion": "0.7.1",
   "skillName": "repo-harness",
   "contractId": "tasks-first-harness-v1",
   "retiredAliases": [
@@ -147,6 +147,10 @@
     {
       "version": "0.7.0",
       "description": "Adds the ChatGPT browser engine with repo-local sessions, opt-in MCP browser tools, an Oracle wrapper, a native Chrome CDP provider spike, and hosted CI gate hardening"
+    },
+    {
+      "version": "0.7.1",
+      "description": "Adds GPT Pro setup and consult command skills over the ChatGPT Web browser bridge and keeps PostBash advisory traces out of the authoritative latest checks trace"
     }
   ],
   "generatedProjectStamp": {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "repo-harness",
-  "version": "0.7.0",
+  "version": "0.7.1",
   "description": "Installs, migrates, audits, and repairs repo-local agentic development harnesses",
   "license": "MIT",
   "repository": {