Borrowing it
Nothing to install: this file belongs to tatargabor/set-core. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/tatargabor/set-core/main/.claude/commands/set/plan-review.mdgit clone --depth 1 https://github.com/tatargabor/set-coreWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/tatargabor/set-core/plan-review)<a href="https://agentmods.dev/commands/tatargabor/set-core/plan-review"><img src="https://agentmods.dev/badge/commands/tatargabor/set-core/plan-review.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00015 | $0.01341 |
| Opus 5 | $0.00008 | $0.00671 |
| Sonnet 5 | $0.00003 | $0.00268 |
| Haiku 4.5 | $0.00002 | $0.00134 |
Grade A, and why
WT: Plan Review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review a spec document against the orchestration planning guide and checklist.
Input: The argument after /set:plan-review is the path to the spec file to review. If no path is given, ask the user which file to review.
Steps
-
Read the planning guide and checklist
- Read
docs/planning-guide.mdfrom the set-core project (find it viawhich set-orchestrate→ resolve to set-core root, or check common locations:~/code2/set-core/docs/,~/code/set-core/docs/) - Read
docs/plan-checklist.mdfrom the same location - If not found locally, check if the current project has copies in its own
docs/directory
- Read
-
Read the spec document
- Read the file path provided by the user
- If no path given, use AskUserQuestion to ask
-
Review against the checklist
Go through each checklist category and evaluate the spec:
Scope & Overlap
- Are scopes self-contained? Any shared files between parallel items?
- Are scopes specific enough (not vague)?
- Any L-sized items that should be split?
Dependencies
- Are schema migrations sequential?
- Is auth foundational?
- Are shared type changes identified?
- Any missing implicit dependencies (barrel exports, config files)?
Manual/Human Tasks
- Do any items involve external services (Stripe, Firebase, AWS, OAuth providers)?
- Do any items require API keys, tokens, or secrets that must be provided by a human?
- Do any items need external account creation, webhook setup, or DNS configuration?
- If yes: flag a warning if the spec doesn't mention manual steps, and suggest adding
[?]tasks with[input:KEY]or[confirm]annotations in the tasks.md
Dependency Graph Analysis Classify each spec item by change type and detect ordering issues:
Change types (in natural execution order):
infrastructure— test/build setup, CI configschema— DB migrations, model changesfoundational— auth, shared types, base componentsfeature— new functionalitycleanup-before— refactor/rename/reorganize that should precede features in same areacleanup-after— dead code removal, cosmetic fixes after features
Check for these ordering issues:
- Cleanup/refactor items alongside feature items that touch the same area → cleanup should run first
- Schema/migration items alongside data-layer or API items → schema should run first
- Auth/role/permission items alongside features that need auth → auth should run first
- Multiple items extending the same type/enum/interface → extract shared change or chain them
- Items with no declared dependencies that clearly depend on each other semantically
Testing
- Does each item mention what to test?
- Is test infrastructure addressed?
- Are existing test patterns referenced?
Sizing & Phases
- Is the batch size reasonable (4-6 per phase)?
- Are phases marked if >6 items?
- Are phases independently valuable?
Directives
- Are orchestrator directives present?
- Is
test_commandset? - Is
merge_policyappropriate?
Design Rules (Layer Check)
- Layer 1 (Business Intent): Is the "what" clear for each item?
- Layer 2 (Constraints): Are boundaries, validation rules, access control specified?
- Layer 3 (Solution Shape): Are selective choices (libraries, UI patterns) reasonable?
- Layer 4 (Implementation): Is anything over-specified that should be left to the agent?
Spec-Writing Checklist per Feature For each feature, check:
- What data does it read/write?
- Who can access it?
- What happens on error?
- Happy path AND edge cases covered?
- Does it interact with existing features?
- What's the UI entry point?
-
Output the review
Format:
## Plan Review: <filename> ### Summary <1-2 sentence overall assessment> ### Checklist Results #### Scope & Overlap - [x] or [ ] for each item, with notes #### Dependencies ... #### Testing ... #### Sizing & Phases ... #### Directives ... ### Per-Feature Review **<Feature 1>** - Layers: ✅ Intent, ✅ Constraints, ⚠️ Missing error handling, ✅ Not over-specified - Data: specified / missing - Access: specified / missing - Tests: specified / missing - Suggestion: <specific improvement> **<Feature 2>** ... ### Dependency Graph **Change Types:** | Item | Type | Notes | |------|------|-------| | <item 1> | feature | — | | <item 2> | cleanup-before | touches same UI area as item 1 | **Suggested Dependencies:** - `<Feature X>`: add `depends_on: <cleanup-change>` (cleanup should precede feature work on shared components) - `<Data layer Y>`: add `depends_on: <schema-change>` (schema migration must complete first) Or if no issues: "Dependency ordering looks good — items are independent (touch different modules/areas)." ### Suggested Improvements 1. <Concrete, actionable suggestion> 2. <Concrete, actionable suggestion> ...
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today First seen · 159 lines · 15 tokens per session scan A 047a454488cc
WT: Plan Review is a command published in the GitHub repository tatargabor/set-core (35 stars, last pushed today), licensed MIT. It adds 15 tokens to every session and 1,341 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-06.
Other commands, from other repositories
hatch3r-pr-resolve
Read open PR comments, evaluate each against current code via the rigor contract, implement accepted findings, reply inline. Multi-platform.
cc-council
Comprehensive multi-agent council review with 6 protocols, 10 specialists, scoped scoring (per-scope thresholds and weights), state machine orchestration, auto-fix, and 50+ configuration flags.
done
Finish a task - document, create PR or merge, close.
deep-review
Multi-pass deep code review — security, quality, and test gaps.
audit
Audit the premises of a document — refute-first, one claim per agent, refutations returned as checkable anchors.
ground
Verify the Evidence Anchors in a document (defaults to the newest plan file).