Base Comparison
stylelint-stylistic/stylelint-stylistic
Measure a branch against the commit it stands on — extract the base instead of flipping the working tree, pick the base by hash, and prove a new test case red on it.
A skill your agent uses for the closing whole-run review in the architect factory — the only model review in the loop: dispatched by the orchestrator, at finish, to one fresh strategist subagent…
Install Final Review
$ npx skills add DanMcInerney/architect-loop --skill final-review -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install DanMcInerney/architect-loop final-review --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/DanMcInerney/architect-loop.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/final-review .claude/skills/final-review && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "final-review" agent skill from https://github.com/DanMcInerney/architect-loop/tree/main/skills/final-review into .claude/skills/final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "final-review", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/DanMcInerney/architect-loop/tree/main/skills/final-reviewType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add DanMcInerney/architect-loop --skill final-review -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install DanMcInerney/architect-loop final-review --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DanMcInerney/architect-loop.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/final-review .agents/skills/final-review && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "final-review" agent skill from https://github.com/DanMcInerney/architect-loop/tree/main/skills/final-review into .agents/skills/final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "final-review", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add DanMcInerney/architect-loop --skill final-review -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install DanMcInerney/architect-loop final-review --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DanMcInerney/architect-loop.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/final-review .cursor/skills/final-review && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "final-review" agent skill from https://github.com/DanMcInerney/architect-loop/tree/main/skills/final-review into .cursor/skills/final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "final-review", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/DanMcInerney/architect-loop.git --path skills/final-review--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add DanMcInerney/architect-loop --skill final-review -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install DanMcInerney/architect-loop final-review --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DanMcInerney/architect-loop.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/final-review .gemini/skills/final-review && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "final-review" agent skill from https://github.com/DanMcInerney/architect-loop/tree/main/skills/final-review into .gemini/skills/final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "final-review", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install DanMcInerney/architect-loop final-reviewInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add DanMcInerney/architect-loop --skill final-review -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/DanMcInerney/architect-loop.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/final-review .github/skills/final-review && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "final-review" agent skill from https://github.com/DanMcInerney/architect-loop/tree/main/skills/final-review into .github/skills/final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "final-review", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add DanMcInerney/architect-loop --skill final-review -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install DanMcInerney/architect-loop final-review --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DanMcInerney/architect-loop.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/final-review .opencode/skills/final-review && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "final-review" agent skill from https://github.com/DanMcInerney/architect-loop/tree/main/skills/final-review into .opencode/skills/final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "final-review", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
final-review- GitHub stars
- 626
- Token cost
- ~1.7k tokens
- SKILL.md length
- 846 words
- Files
- 2
- Skills in repo
- 11
- Repo updated
- First seen
- Licence
- MIT
At a glance
A skill your agent uses for the closing whole-run review in the architect factory — the only model review in the loop: dispatched by the orchestrator, at finish, to one fresh strategist subagent…
- Works in 4 steps: The spec (docs/spec/.md): goal,… → The full run diff: git diff ..HEAD. → Every shipped issue's published… → …
- The closing whole-run review in the architect factory — the only model review in the loop: dispatched by the orchestrator
- SKILL.md covers Review basis, in order, Gates on every finding, Cohesion and Spec, plus 4 more sections
- Calls git
What it does
Final Review is an agent skill from DanMcInerney/architect-loop. Use for the closing whole-run review in the architect factory — the only model review in the loop: dispatched by the orchestrator, at finish, to one fresh strategist subagent that audits the entire run diff for defects only isolated parallel slices can produce, checks the merged whole against the spec, verifies every candidate finding, and delivers a review spec plus draft fix issues and draft graded checks — it never edits product code or the mutable test suite. Never description-triggered or self-invoked…
Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `TEST-STEWARDSHIP.md`).
It sits in Agent Workflows, covering Test generation and Subagents. The repository describes itself as: Super optimized /goal loop. Massive token savings and higher quality. Smart model designs and reviews, cheaper model builds.. The licence is MIT.
When your agent uses it
- The closing whole-run review in the architect factory — the only model review in the loop: dispatched by the orchestrator
- One fresh strategist subagent that audits the entire run diff for defects only isolated parallel slices can produce
- Checks the merged whole against the spec
- Verifies every candidate finding
Example prompts
- “/final-review”
Workflow steps
4 steps, taken from the first numbered list in SKILL.md.
- The spec (docs/spec/.md): goal, non-goals, validation strategy.
- The full run diff: git diff ..HEAD.
- Every shipped issue's published interface contract block.
- The closing test-pass output in your dispatch block — builder-built suites plus every frozen RUN item, run by the orchestrator at the head…
What it can do on your machine
Read from SKILL.md and the folder at commit 28dca7d. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
git
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Final Review loads about 1.7k tokens when it runs. Until then it costs about 150 tokens; SKILL.md has 846 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from DanMcInerney/architect-loop at commit 28dca7d, republished under its MIT licence (© DanMcInerney). 846 words, ~1,674 tokens.
Download SKILL.mdSave it as
.claude/skills/final-review/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.- name
- final-review
- description
- Use for the closing whole-run review in the architect factory — the only model review in the loop: dispatched by the orchestrator, at finish, to one fresh strategist subagent that audits the entire run diff for defects only isolated parallel slices can produce, checks the merged whole against the spec, verifies every candidate finding, and delivers a review spec plus draft fix issues and draft graded checks — it never edits product code or the mutable test suite. Never description-triggered or self-invoked mid-run; the orchestrator calls it explicitly once every issue has closed.
<!-- Adapted from mattpocock/skills (MIT). -->Final Review
You review a finished run cold — you built nothing here; that's the point. Fresh context catches what builders inside their own worktrees could not see. You are the only model review in the loop: builders ran their own tests and the check-runner graded the frozen checks; everything a fresh reader can catch lands on you. You review and decompose; you never edit — your findings ship through a fix wave of fresh builders the check-runner grades, same as any other issue.
Review basis, in order
- The spec (
docs/spec/<run>.md): goal, non-goals, validation strategy. - The full run diff:
git diff <pre-run-sha>..HEAD. - Every shipped issue's published interface contract block.
- The closing test-pass output in your dispatch block — builder-built suites plus every frozen RUN item, run by the orchestrator at the head you review.
Dispatch mechanics — worktree from the factory branch head, docs/checks/
read-only, you commit nothing, downstream harvest/freeze/filing/dispatch is
the orchestrator's — follow skills/architect/SKILL.md ### 5. Finish as
given; this skill does not restate those mechanics.
Gates on every finding
- Scope: only defects introduced by this run's diff. A pre-existing issue gets one digest line, never a fix. [O-SCOPE]
- Confidence: "If you are not certain an issue is real, do not flag it." Prefer no findings over weak findings. [A-CONF][O-PREF]
- Verify, then report: reproduce each candidate BEFORE writing it up — run the code path, or demonstrate the contradiction with file:line pairs. Candidates you cannot reproduce are dropped, not reported. [A-VAL]
Cohesion
Isolation is what let slices run in parallel; it is also the only thing that can go wrong here. Walk the diff hunting for:
- Duplicated concepts or helpers implemented twice under different names.
- Naming that diverges from the codebase-design glossary (below).
- Interface drift: a producer slice's published contract vs. what its consumers actually call.
- Contradictory cross-slice assumptions — A assumes absent what B added; C removes what D extends.
- Inconsistent error handling for the same class of failure across slices.
- Shared-surface tracing: walk every surface two or more slices touch and confirm both edits agree on its shape.
- Stale or superseded code the run left behind, and any backwards-compatibility shim the spec never asked for — report both as findings; the factory keeps no unrequested compat code.
Spec
Set the diff down and reread the spec's goal, non-goals, and validation strategy. Report: requirements that are missing or partial; behavior the spec never asked for (scope creep); requirements that look implemented but where the implementation looks wrong.
Reporting
Grade each verified finding: P0 breaks the run's goal or checks; P1 wrong behavior against the spec; P2 cohesion debt. [O-SEV] One short paragraph per finding — the explicit scenario in which it fails, matter-of-fact tone, code excerpts of at most 3 lines. [O-FMT]
Do not merge or rerank findings — the two axes are deliberately separate. Report total findings, severity counts, and the worst finding within each axis; never a single winner across axes.
Calibration: flag only gaps that affect correctness, the stated requirements, or documented project invariants — cite file:line evidence; no stylistic preferences.
End your final message with exactly one verdict line. Zero verified
findings: REVIEW: GREEN — no review spec, no drafts, the verdict line is
the whole report. One or more verified findings: REVIEW: FINDINGS n=<count>
followed by the draft locations from ## Decompose discipline below.
Severity counts and the per-axis worst finding accompany either line.
Show full SKILL.md (294 more words)Show less
Decompose discipline
One or more verified findings: write the review spec at
docs/runs/<run>/review-spec.md — one requirement per finding, each
carrying its severity and the file:line verification from the gates above.
A run artifact, not the run's hardened spec.
Cut it into fix issues per the to-issues discipline
(skills/to-issues/SKILL.md): tracer-bullet slices, structural before
behavioral, the disjoint parallel frontier, blocked-by edges, published
interface contracts, a change-skeleton per issue. Draft each at
docs/runs/<run>/review/issues/<slug>.md.
Draft one graded check per fix issue per the frozen-checks discipline
(skills/frozen-checks/SKILL.md) — purpose, spec pointer, fix contract,
falsifiable RUN items run against the current tree — at
docs/runs/<run>/review/checks/<slug>.md.
You write drafts only: no draft authorizes an edit from you to product code
or the test suite — the fix wave's builders make those edits, graded by the
check-runner. You commit nothing, never touch docs/checks/, and never
file or otherwise mutate the tracker; the orchestrator harvests, rules on,
freezes, and files these drafts.
Test stewardship
Your scope includes the run's mutable test suite, as diagnosis, not edits:
map spec behaviors to tests at their seam per TEST-STEWARDSHIP.md. Gaps,
misclassified tests, and unfalsifiable tests are findings; each becomes a
fix issue carrying the falsifiability proof (an add) or the classified
reason (a rewrite or deletion) — you execute no test edits yourself.
Frozen checks under docs/checks/ are a separate immutable layer — never
edited, never a substitute for the mutable suite; every graded RUN item
stays green after the fix wave's test edits.
Glossary contract
Use the codebase-design glossary (skills/codebase-design/SKILL.md)
exactly: module, interface,
implementation, seam, adapter, depth, leverage, locality; run, tracking
issue, issue, slice, frozen check, check-runner, strategist, builder,
orchestrator, factory branch, worktree, job report, verdict, ruling, digest,
hard stop. Do not substitute component/service/boundary/API for
module/interface, or task/ticket for issue — a substitution is itself a
cohesion finding, not a style choice.
© DanMcInerney, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 1 other file in skills/final-review of DanMcInerney/architect-loop.
- SKILL.md
- TEST-STEWARDSHIP.md
Open the folder on GitHubat commit 28dca7d
Compare with similar skills
Final Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Final Review this skillDanMcInerney/architect-loop | 626 | — | ~1.7k | Automated safety check: Pass | MIT | |
| Base Comparisonstylelint-stylistic/stylelint-stylistic | 106 | — | ~2.2k | Automated safety check: Pass | Custom licence | |
| User Reviewantoinecellerier/speaker-tuning-to-easyeffects | 143 | — | ~4.3k | Automated safety check: Pass | MIT | |
| Poweruser Feature Auditpydantic/pydantic-ai | 20k | — | ~2.9k | Automated safety check: Pass | MIT | |
| Parallel Test Fixingspencerpauly/awesome-cursor-skills | 843 | — | ~525 | Automated safety check: Pass | CC0-1.0 | |
| Bench Batonnooga/let-go | 568 | — | ~821 | Automated safety check: Pass | MIT |
Similar skills
User Review
antoinecellerier/speaker-tuning-to-easyeffects
Reviews the scripts' user-facing terminal output by running it past subagent reviewers role-playing a first-time user, then reports severity-ranked findings.
- Official
Poweruser Feature Audit
pydantic/pydantic-ai
Independent power-user audit of a big new-feature PR. An agent skill from pydantic/pydantic-ai.
Parallel Test Fixing
spencerpauly/awesome-cursor-skills
When multiple tests fail, assign each failing test file to a separate subagent that fixes it independently in parallel.
Bench Baton
nooga/let-go
Coordinate heavy local workloads across worktrees, processes, and subagents — benchmarks and timing-sensitive gates run exclusively on a quiesced machine, while builds, test suites, regeneration…
Antigravity
yuting0624/antigravity-for-claude-code
Run the Antigravity CLI (Gemini) as a collaborating AI inside Claude Code, with intelligent model routing across the software development lifecycle.
More from DanMcInerney/architect-loop
All 11 skills in this repoArchitect Research
DanMcInerney/architect-loop
A skill your agent uses when the user asks for discovery-scale research that informs a decision: brainstorming a project or feature, choosing a technology, or requests like "research X", "what's the…
Adversarial Review
DanMcInerney/architect-loop
A skill your agent uses when the architect factory orchestrator dispatches a fresh strategist subagent to harden a draft spec: falsify it with file:line evidence, fold the surviving findings into a…
Architect
DanMcInerney/architect-loop
A skill your agent uses when the user asks to architect, run or continue the autonomous software factory, turn a goal into a hardened tracker issue plan, dispatch builder jobs, grade finished work…
Architect Fast
DanMcInerney/architect-loop
A skill your agent uses when the user asks to architect-fast a change, run the light factory lane, or factory-build a small goal — a few files, roughly one sitting, at most ~3 parallel issues — into…
Frozen Checks
DanMcInerney/architect-loop
A skill your agent uses when the strategist drafts per-issue graded checks after decomposition and before builder dispatch.
TDD
DanMcInerney/architect-loop
Test-driven development for factory builders. An agent skill from DanMcInerney/architect-loop.
Related
Categories
Questions about Final Review
What does Final Review do?
A skill your agent uses for the closing whole-run review in the architect factory — the only model review in the loop: dispatched by the orchestrator, at finish, to one fresh strategist subagent…. Final Review is an agent skill from DanMcInerney/architect-loop. Use for the closing whole-run review in the architect factory — the only model review in the loop: dispatched by the orchestrator, at finish, to one fresh strategist subagent that audits the entire run diff for defects only isolated parallel slices can produce, checks the merged whole against the spec, verifies every candidate finding, and delivers a review spec plus draft fix issues and draft graded checks — it never edits product code or the mutable test suite.
When should I use Final Review?
Final Review fits situations like: the closing whole-run review in the architect factory — the only model review in the loop: dispatched by the orchestrator; one fresh strategist subagent that audits the entire run diff for defects only isolated parallel slices can produce; checks the merged whole against the spec; verifies every candidate finding.
How do I install Final Review in Claude Code?
Run `npx skills add DanMcInerney/architect-loop --skill final-review -a claude-code`. Or copy the skill folder (skills/final-review in DanMcInerney/architect-loop) into .claude/skills/final-review in your project. Claude Code loads it when a task matches its description.
How do I install Final Review in Codex?
Run `npx skills add DanMcInerney/architect-loop --skill final-review -a codex`. Or copy the skill folder (skills/final-review in DanMcInerney/architect-loop) into .agents/skills/final-review in your project. Codex loads it when a task matches its description.
Can I use Final Review in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add DanMcInerney/architect-loop --skill final-review -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/final-review, .gemini/skills/final-review, .github/skills/final-review and .opencode/skills/final-review in your project.
What does Final Review need to run?
Going by SKILL.md and its folder, Final Review needs the command-line tools its instructions call (git).
Does Final Review access the network?
SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Final Review safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Final Review use?
Final Review is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Final Review use?
About 1.7k tokens (SKILL.md is roughly 6.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Final Review?
Skills that share tags, products or a category with Final Review: Base Comparison (stylelint-stylistic/stylelint-stylistic, 106 stars), User Review (antoinecellerier/speaker-tuning-to-easyeffects, 143 stars), Poweruser Feature Audit (pydantic/pydantic-ai, 20k stars) and Parallel Test Fixing (spencerpauly/awesome-cursor-skills, 843 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Final Review?
DanMcInerney (a GitHub user) maintains it in DanMcInerney/architect-loop, which has 626 GitHub stars. The repository holds 11 skills in this directory. The repository was last updated on September 13, 2026.
Source: DanMcInerney/architect-loop on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.