plan-review-handoff-verification
Verify a parked GitHub issue plan-review handoff across live labels, local markers, README rows, canonical review artifacts, and stale plan-body evidence before reporting next action.
Best use case
plan-review-handoff-verification is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Verify a parked GitHub issue plan-review handoff across live labels, local markers, README rows, canonical review artifacts, and stale plan-body evidence before reporting next action.
Teams using plan-review-handoff-verification should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/plan-review-handoff-verification/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How plan-review-handoff-verification Compares
| Feature / Agent | plan-review-handoff-verification | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Verify a parked GitHub issue plan-review handoff across live labels, local markers, README rows, canonical review artifacts, and stale plan-body evidence before reporting next action.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Plan-review handoff verification
Use this when a user provides or references an exit handoff for a GitHub issue parked in `status:plan-review`, especially after iterative adversarial review and approval-state drift cleanup.
## Workflow
1. Load the canonical planning skill (`issue-planning-mode`) first.
2. Read the handoff file and extract the asserted state surfaces:
- GitHub issue state and `status:*` labels
- `.planning/plan-approved/<issue>.md` marker state
- `docs/plans/README.md` row status
- canonical provider review artifact paths and verdicts
- stated clean next action / do-not-do guidance
3. Verify the live GitHub state with `gh issue view <issue> --json state,labels,title,url`.
4. Verify the local approval marker presence/absence.
5. Search `docs/plans/README.md` for the issue row and confirm the row status and note match the intended parked state.
6. Read or scan the canonical `scripts/review/results/*plan-<issue>-<provider>.md` artifacts and extract each verdict.
7. Read the local plan header and scan the plan body for stale older rerun evidence (for example a review-artifacts line still saying `r11 MAJOR` while canonical artifacts now say `r16 MINOR`).
8. Report the operational state and next valid action. Do not implement, approve, recreate approval markers, or rewrite broad plan content unless the user explicitly asks.
## Classification rules
- Operational surfaces are, in order: live GitHub status labels, local approval marker, `docs/plans/README.md` row, canonical review artifacts.
- If those surfaces align to `status:plan-review` with no approval marker and no fresh MAJOR verdicts, classify the issue as parked/exit-clean for user review.
- If the local plan header/body contains stale historical evidence while the operational surfaces are aligned, report it as non-blocking polish/drift. Do not treat it as approval or implementation authority.
- If a stale approval marker or `status:plan-approved` label remains after fresh review evidence says the issue is back in review, perform the governance cleanup from `issue-planning-mode` instead of merely noting it.
## Minimal verification command pattern
Use one script to reduce missed surfaces:
```python
from hermes_tools import terminal, search_files, read_file
import os, json, re
issue = 2460
state = terminal(f"gh issue view {issue} --json state,labels,title,url", timeout=60)
marker_exists = os.path.exists(f".planning/plan-approved/{issue}.md")
readme = search_files(str(issue), target="content", path="docs/plans/README.md", output_mode="content")
for provider in ["Codex", "codex", "gemini"]:
path = f"scripts/review/results/2026-04-23-plan-{issue}-{provider}.md"
# read_file(path) and extract Verdict: APPROVE|MINOR|MAJOR
```
Then separately scan the plan file for stale older strings such as `MAJOR`, `UNAVAILABLE`, or older rerun IDs when the latest artifacts have advanced.
## Output pattern
State the result as:
1. GitHub issue state and live status labels
2. Local approval marker state
3. README row state
4. Provider artifact verdicts
5. Local plan file/header state
6. Conclusion: exit-clean vs drift found
7. Next valid action and prohibited action
Keep the response short and explicit: if parked for user approval, say no implementation should start yet.Related Skills
subagent-write-verification
Independently verify subagent-claimed file writes with filesystem and git checks before treating the artifact as real, before committing it, and before referencing the path in downstream prompts.
handoff
Compact the current conversation into a handoff document for another agent to pick up.
context-compaction-handoff
Guardrails for resuming work after context compaction or transcript handoff blocks; prioritize the latest real user request over stale summarized tasks and verify before answering.
plan-review-prompt-refresh-after-plan-edits
Refresh reviewer prompt files from the latest on-disk plan before every adversarial re-review. Prevents Codex/Gemini from critiquing stale plan text after local edits.
workspace-hub-overnight-plan-monitor
Monitor and reconcile workspace-hub overnight planning or implementation batches, including process status, result artifacts, issue/commit verification, and controlled failed-lane recovery.
plan-gated-issue-execution-wave
Execute a multi-issue architecture/planning wave in a plan-gated repo, then safely transition approved issues into implementation with file-based Codex prompts, local approval markers, subprocess monitoring, and cleanup handling for sandbox/hook edge cases.
mixed-ops-vs-repo-fix-plan-boundary
Plan mixed operational-vs-repo remediation issues by proving live-state classification first, then only proposing code changes for confirmed repo-owned failure paths.
wave-based-parallel-plan-execution
Orchestrate phase execution by discovering dependencies, grouping into waves, spawning subagents, and collecting results with optional wave filtering
tdd-verification-and-adversarial-review
Verify pre-written TDD tests pass, conduct adversarial code review on committed diffs, and route findings to existing issues
tax-form-navigation-verification
Systematic workflow for verifying tax software calculations against entry guide specifications
portable-pattern-verification-workflow
Multi-package implementation with verification strategy for cross-platform configuration hardening
plan-gated-issue-validation-workflow
Systematic validation pattern for plan-approved GitHub issues with pre-existing deliverables