Skip to content
Agentmart
Submit a tool

hitl-review

Manage the Human-In-The-Loop review queue for tests with uncertain verdicts (confidence 0.7–0.9).

Skill by moaidmoatasem

IndexedApache-2.0★ 1
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/moaidmoatasem-hitl-review

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Manage the Human-In-The-Loop review queue for tests with uncertain verdicts (confidence 0.7–0.9).

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/hitl-review

What's inside

Skill name
hitl-review
Description
Manage the Human-In-The-Loop review queue for tests with uncertain verdicts (confidence 0.7–0.9).
License
Apache-2.0
Instructions
72 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# HITL Review Skill

## Purpose
Manage the Human-In-The-Loop review queue when the REVIEW stage produces uncertain verdicts (confidence 0.7-0.9). Items can be triaged via CLI or Web UI.

## When to Use
- A REVIEW stage returns `Verdict.HITL` for a generated test
- A human needs to decide: approve (keep), reject (discard), or classify (regression/intended/ignore)
- You want AI explanation of why an item was flagged

## Workflow

### CLI Operations (`cherenkov/hitl/cmd.py`)

```bash
# List pending items
./bin/cherenkov hitl list
./bin/cherenkov hitl list --all
./bin/cherenkov hitl list --status approved --json

# Inspect a specific item
./bin/cherenkov hitl show <item_id>

# Approve (atomic — only one actor wins on race)
./bin/cherenkov hitl approve <item_id> --actor @alice

# Reject with required reason
./bin/cherenkov hitl reject <item_id> --reason "incorrect_spec"

# Classify (Tier-2)
./bin/cherenkov hitl classify <item_id> --classification regression

# AI explanation (Tier-3)
./bin/cherenkov hitl explain <item_id>
```

All subcommands support `--json` for `hitl/v1` envelope output.

### Web UI Operations (`cherenkov/web/api.py`)

```bash
./bin/cherenkov dashboard
```

Opens a browser-based review dashboard with:
- Findings queue (approve/reject/classify buttons)
- "Why flagged?" AI explanation panel
- Rejection reason capture
- No build step — ships prebuilt `dist/`

### SQLite Queue (`cherenkov/hitl/store.py`)
- Durable SQLite store at `.cherenkov/hitl.db`
- Atomic status transitions: `pending → approved | rejected`
- Race-condition safe (SQLite write-lock)

### Error Codes
| Code | Meaning |
|------|---------|
| `conflict` | Item already resolved by another actor |
| `not_found` | Item ID not found |
| `forbidden` | Actor not authorized |
| `db_locked` | SQLite busy-timeout exceeded |

## References
- `cherenkov/hitl/cmd.py` — CLI hitl dispatch
- `cherenkov/hitl/store.py` — SQLite queue implementation
- `cherenkov/web/api.py` — FastAPI review endpoints
- `cherenkov/web/ui/` — React/Vite frontend
- `smoke_test_hitl_cli.py`, `smoke_test_hitl_race.py`, `smoke_test_hitl_concurrency.py` — smoke tests
- `docs/GETTING_STARTED.md` — user-facing HITL docs

From moaidmoatasem/cherenkov-qa, Apache-2.0.

More from moaidmoatasem/cherenkov-qa

Signals

Stars
1
Forks
0
Open issues
24
Last commit (this folder)
Aug 8, 2026 (2 months ago)
Last release
Aug 2, 2026 (2 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from moaidmoatasem

Cite this page

hitl-review skill. Agent Mart. https://agentmart.dev/skills/moaidmoatasem-hitl-review (last updated 2026-10-24)