Skip to content

Commit 338b4dd

Browse files
committed
Remove PR loop feature entirely
The PR loop workflow is superseded by the /loop command combined with GitHub PR review polling. This removes all PR loop implementation, tests, documentation, templates, and supporting scripts. Deleted PR-loop-only files: commands (start/cancel-pr-loop), hooks (pr-loop-stop-hook), scripts (setup/cancel-pr-loop, check-bot-reactions, check-pr-reviewer-status, fetch-pr-comments, poll-pr-reviews), prompt templates (pr-loop/), test fixtures (setup-fixture-mock-gh), and all PR loop test files. Cleaned PR loop references from mixed files: hook validators, loop library, template loader, bash validator, humanize.sh monitor, monitor-common.sh, codex hooks config, install script, RLCR setup mutual exclusion, SKILL.md, usage docs, and test harnesses. The install script retains pr-loop-stop-hook.sh in its managed pattern so upgrading users get stale hooks cleaned from their hooks.json.
1 parent 4561bc4 commit 338b4dd

61 files changed

Lines changed: 32 additions & 13483 deletions

File tree

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

.gitignore

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,7 @@ temp
33

44
# Local Claude client settings
55
/.claude/settings.json
6+
/.claude/scheduled_tasks.lock
67

78
# Humanize state directories (runtime-generated, project-local)
89
.humanize/

commands/cancel-pr-loop.md

Lines changed: 0 additions & 25 deletions
This file was deleted.

commands/start-pr-loop.md

Lines changed: 0 additions & 61 deletions
This file was deleted.

config/codex-hooks.json

Lines changed: 1 addition & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
{
2-
"description": "Humanize Codex Hooks - Native Stop hooks for RLCR and PR loops",
2+
"description": "Humanize Codex Hooks - Native Stop hooks for RLCR loops",
33
"hooks": {
44
"Stop": [
55
{
@@ -9,12 +9,6 @@
99
"command": "{{HUMANIZE_RUNTIME_ROOT}}/hooks/loop-codex-stop-hook.sh",
1010
"timeout": 7200,
1111
"statusMessage": "humanize RLCR stop hook"
12-
},
13-
{
14-
"type": "command",
15-
"command": "{{HUMANIZE_RUNTIME_ROOT}}/hooks/pr-loop-stop-hook.sh",
16-
"timeout": 7200,
17-
"statusMessage": "humanize PR stop hook"
1812
}
1913
]
2014
}

docs/install-for-claude.md

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -68,7 +68,6 @@ Then use:
6868

6969
```bash
7070
humanize monitor rlcr # Monitor RLCR loop
71-
humanize monitor pr # Monitor PR loop
7271
```
7372

7473
## Other Install Guides

docs/install-for-codex.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -76,7 +76,7 @@ sed -n '1,220p' "${CODEX_HOME:-$HOME/.codex}/hooks.json"
7676

7777
Expected:
7878
- `codex_hooks` is `true`
79-
- `hooks.json` contains `loop-codex-stop-hook.sh` and `pr-loop-stop-hook.sh`
79+
- `hooks.json` contains `loop-codex-stop-hook.sh`
8080
- `${XDG_CONFIG_HOME:-~/.config}/humanize/config.json` contains `bitlesson_model` set to a Codex/OpenAI model such as `gpt-5.4`
8181
- for `--target codex`, `${XDG_CONFIG_HOME:-~/.config}/humanize/config.json` also contains `provider_mode: "codex-only"`
8282

docs/usage.md

Lines changed: 2 additions & 40 deletions
Original file line numberDiff line numberDiff line change
@@ -63,8 +63,6 @@ The quiz is advisory, not a gate. You always have the option to proceed. But tha
6363
| `/cancel-rlcr-loop` | Cancel active loop |
6464
| `/gen-plan --input <draft.md> --output <plan.md>` | Generate structured plan from draft |
6565
| `/refine-plan --input <annotated-plan.md>` | Refine an annotated plan and generate a QA ledger |
66-
| `/start-pr-loop --claude\|--codex` | Start PR review loop with bot monitoring |
67-
| `/cancel-pr-loop` | Cancel active PR loop |
6866
| `/ask-codex [question]` | One-shot consultation with Codex |
6967

7068
## Command Reference
@@ -209,39 +207,6 @@ If `--alt-language` is set to a supported non-English language, the command also
209207
translated plan and QA variants by inserting `_<code>` before the file extension, such as
210208
`plan_zh.md` and `plan-qa_zh.md`.
211209

212-
### start-pr-loop
213-
214-
```
215-
/humanize:start-pr-loop --claude|--codex [OPTIONS]
216-
217-
BOT FLAGS (at least one required):
218-
--claude Monitor reviews from claude[bot] (trigger with @claude)
219-
--codex Monitor reviews from chatgpt-codex-connector[bot] (trigger with @codex)
220-
221-
OPTIONS:
222-
--max <N> Maximum iterations before auto-stop (default: 42)
223-
--codex-model <MODEL:EFFORT>
224-
Codex model and reasoning effort (default from config, effort: medium)
225-
--codex-timeout <SECONDS>
226-
Timeout for each Codex review in seconds (default: 900)
227-
-h, --help Show help message
228-
```
229-
230-
The PR loop automates the process of handling GitHub PR reviews from remote bots:
231-
232-
1. Detects the PR associated with the current branch
233-
2. Fetches review comments from the specified bot(s)
234-
3. Claude analyzes and fixes issues identified by the bot(s)
235-
4. Pushes changes and triggers re-review by commenting @bot
236-
5. Stop Hook polls for new bot reviews (every 30s, 15min timeout per bot)
237-
6. Local Codex validates if remote concerns are approved or have issues
238-
7. Loop continues until all bots approve or max iterations reached
239-
240-
**Prerequisites:**
241-
- GitHub CLI (`gh`) must be installed and authenticated
242-
- Codex CLI must be installed
243-
- Current branch must have an associated open PR
244-
245210
### ask-codex
246211

247212
```
@@ -284,7 +249,7 @@ Current built-in keys:
284249

285250
### Codex Model Configuration
286251

287-
All Codex-using features (RLCR loop, PR loop, ask-codex) share the same model configuration:
252+
All Codex-using features (RLCR loop, ask-codex) share the same model configuration:
288253

289254
| Key | Default | Description |
290255
|-----|---------|-------------|
@@ -307,7 +272,7 @@ are unset, so BitLesson selection stays on the Codex/OpenAI path without probing
307272

308273
Codex model is resolved with this precedence:
309274
1. CLI `--codex-model` flag (highest priority)
310-
2. Feature-specific defaults (e.g., PR loop defaults to `medium` effort)
275+
2. Feature-specific defaults
311276
3. Config-backed defaults from the 4-layer hierarchy above
312277
4. Hardcoded fallback (`gpt-5.4:high`)
313278

@@ -327,16 +292,13 @@ source ~/.claude/plugins/cache/humania/humanize/<LATEST.VERSION>/scripts/humaniz
327292
# Monitor RLCR loop progress
328293
humanize monitor rlcr
329294

330-
# Monitor PR loop progress
331-
humanize monitor pr
332295
```
333296

334297
Progress data is stored in `.humanize/rlcr/<timestamp>/` for each loop session.
335298

336299
## Cancellation
337300

338301
- **RLCR loop**: `/humanize:cancel-rlcr-loop`
339-
- **PR loop**: `/humanize:cancel-pr-loop`
340302

341303
## Environment Variables
342304

hooks/hooks.json

Lines changed: 1 addition & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
{
2-
"description": "Humanize Plugin Hooks - Validation hooks and Stop hooks for /start-rlcr-loop and /start-pr-loop",
2+
"description": "Humanize Plugin Hooks - Validation hooks and Stop hooks for /start-rlcr-loop",
33
"hooks": {
44
"UserPromptSubmit": [
55
{
@@ -67,11 +67,6 @@
6767
"type": "command",
6868
"command": "${CLAUDE_PLUGIN_ROOT}/hooks/loop-codex-stop-hook.sh",
6969
"timeout": 7200
70-
},
71-
{
72-
"type": "command",
73-
"command": "${CLAUDE_PLUGIN_ROOT}/hooks/pr-loop-stop-hook.sh",
74-
"timeout": 7200
7570
}
7671
]
7772
}

0 commit comments

Comments
 (0)