review-deep

review-deep is a skill for Claude Code from juliusz-cwiakalski/agentic-delivery-os. It costs 26 tokens per session (1,595 once invoked), scanned A, original, MIT.

A review workflow for comparing a code change with its specification, plan, project rules, and code-quality expectations. It can add a remediation phase when problems are found.

In plain words
What is it for?
Use it to review a planned change, identify its specification and branch, compare its differences with a base branch, and check implementation quality and required follow-up work.
Why use it?
It provides a structured way to find gaps between intended and delivered work, then track fixes against the same change.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: model in frontmatter; positional $N argument; mentions AGENTS.md.

Part of the ados plugin — 20 skills, 24 agents shipped together

Good fit Use it to review a planned change, identify its specification and branch, compare its differences with a base branch, and check implementation quality and required follow-up work.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/juliusz-cwiakalski/agentic-delivery-os/review-deep
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add juliusz-cwiakalski/agentic-delivery-os --skill review-deep
Clone the repo
git clone --depth 1 https://github.com/juliusz-cwiakalski/agentic-delivery-os

Made for: Claude Code.

Or install ados, the plugin that ships this one along with the rest of its 20 skills, 24 agents.

Wrote 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.

agentmods badge for review-deep

README.md
[![agentmods](https://agentmods.dev/badge/skills/juliusz-cwiakalski/agentic-delivery-os/review-deep/github.svg)](https://agentmods.dev/skills/juliusz-cwiakalski/agentic-delivery-os/review-deep)
Your own site
<a href="https://agentmods.dev/skills/juliusz-cwiakalski/agentic-delivery-os/review-deep"><img src="https://agentmods.dev/badge/skills/juliusz-cwiakalski/agentic-delivery-os/review-deep/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for review-deep

Your own site · 80×15
<a href="https://agentmods.dev/skills/juliusz-cwiakalski/agentic-delivery-os/review-deep"><img src="https://agentmods.dev/badge/skills/juliusz-cwiakalski/agentic-delivery-os/review-deep.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 26 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,595 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Excessive Agency · line 11
    Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.
    Fix: Remove the model/provider override or disclose it prominently and require explicit operator approval before invoking an external coding CLI or billed model.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00026 $0.01595
Opus 5 $0.00013 $0.00797
Sonnet 5 $0.00005 $0.00319
Haiku 4.5 $0.00003 $0.00160

Measured 12d ago against content hash c5d79484d9bd, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

review-deep 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 12d ago.

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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

  • review — 91% identical, 20 lines differ
.ados-claude/skills/review-deep/SKILL.md · 157 lines

How it starts

The opening of the file, as written. The whole thing — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.

<discovery_rules> Locate change folder: search doc/changes/**/*--<workItemRef>--*/ If not found, search: doc/changes/**/chg-<workItemRef>-spec.md Spec file: chg-<workItemRef>-spec.md; derive slug & change.type from frontmatter. Plan file: chg-<workItemRef>-plan.md Folder pattern: doc/changes/YYYY-MM/YYYY-MM-DD--<workItemRef>--<slug>/ Abort with clear error if spec OR plan missing. </discovery_rules>

<branch_resolution> changeBranch = <change.type>/<workItemRef>/<slug> headRef: directives override, else try changeBranch, fallback: current HEAD. baseBranch: directives override, else main, fallback: master. Compute merge-base and gather symmetric diff baseBranch...headRef. </branch_resolution>

<directive_parsing> Directives (case-insensitive):

  • Base branch: base=<branch> | base branch <branch> | compare vs <branch>
  • Head ref: head=<ref> | head ref <ref> | branch <ref>
  • Suppress commit trigger: no commit | commit=false
  • Dry run: dry run | preview only Unrecognized tokens ignored. </directive_parsing>

<pre_flight>

  1. Validate workItemRef format (uppercase prefix + hyphen + digits).
  2. Resolve spec; extract change.type & slug.
  3. Resolve plan; parse phases & existing remediation.
  4. Load repository rules (AGENTS.md, .ai/rules/**).
  5. Resolve branches & compute diff metadata.
  6. If dryRun: prepare preview but do NOT write. </pre_flight>

<review_method> The reviewer agent applies its full review framework:

Spec/plan compliance (local mode):

  • Scope compliance: changed files align with spec capabilities
  • Plan alignment: all tasks done, acceptance criteria have evidence
  • Plan task audit: OPEN_TASKS, DONE_BUT_UNCHECKED, CHECKED_BUT_MISSING
  • Out-of-scope detection: changes to files not in plan

Code quality heuristics (both modes):

  • Full built-in heuristic framework: correctness, security, performance, reliability, API compat, testing gaps, documentation, dependencies
  • Repository-local rules from .ai/agent/code-review-instructions.md and .ai/rules/

Ticket context (when available):

  • Acceptance criteria verification against implementation
  • Linked issue traversal for additional constraints and decisions

The reviewer loads all available context — spec, plan, ticket, repo rules, heuristics — and applies them holistically. Do not duplicate heuristic definitions here; they live in the reviewer agent prompt. </review_method>

<findings_format> [severity: major|minor|nit] <file>[:line] — <description>; fix: <action> </findings_format>

<remediation_phase> If findings exist, append new phase to plan:

### Phase X: Code Review Remediation

- Goal: Address code review findings.
- Tasks:
  - [ ] <precise fix per finding>
- Acceptance criteria:
  - Must: All fixes implemented and validated.
  - Must: Updated tests pass.
- Files and modules: <paths>
- Completion signal: docs(plan): remediate review findings for <workItemRef>

Rules:

  • X = max existing phase + 1.
  • Do NOT modify earlier phases.
  • Append revision log entry. </remediation_phase>

Read the full file on GitHub · 157 lines

Changes

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.

  1. 12d ago First seen · 157 lines · 26 tokens per session scan A c5d79484d9bd

Subscribe to this mod's changes

review-deep is a skill published in the GitHub repository juliusz-cwiakalski/agentic-delivery-os (38 stars, last pushed 2d ago), licensed MIT. It adds 26 tokens to every session and 1,595 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 2026-08-30.

Related

Other skills, from other repositories

review

Review before merge. Stage-1 spec-compliance gate, then risk-selected Stage-2 review axes from the canonical set. analyst always runs, callers can pin extra always-on axes, and explicit deep review runs the full 16-axis set. Run after /test. Do NOT invoke code-qualities-assessment, doc-accuracy, golden-principles, or…

rjmurillo/ai-agents · 97 tokens

code-qualities-assessment

Assess code maintainability through 5 foundational qualities (cohesion, coupling, encapsulation, testability, non-redundancy) with quantifiable scoring rubrics. Works at method/class/module levels across multiple languages. Produces markdown reports with remediation guidance. Use when you ask to "assess…

rjmurillo/ai-agents · 108 tokens

adr-review

Multi-agent debate orchestration for Architecture Decision Records. Automatically triggers on ADR create/edit/delete. Coordinates architect, critic, independent-thinker, security, analyst, and high-level-advisor agents in structured debate rounds until consensus. Use when you say "review this ADR", when an ADR is…

rjmurillo/ai-agents · 134 tokens

style-enforcement

Validate code against style rules from .editorconfig, StyleCop.json, and Directory.Build.props. Use when you say "check style compliance", "validate editorconfig rules", "run style enforcement", "check naming conventions", or "validate line endings". Detects line ending violations, naming convention issues…

rjmurillo/ai-agents · 103 tokens

analysis-provenance

Identify code ownership before modifying validators or linters. Checks file headers for provenance indicators, reviews documentation, and determines provenance as UPSTREAM, LOCAL, VENDOR, or UNKNOWN. Prevents accidental modification of upstream tools.

rjmurillo/ai-agents · 48 tokens

analyze

Systematic multi-step codebase analysis producing prioritized findings with file-line evidence. Covers architecture reviews, security assessments, and code quality evaluations through guided exploration, investigation planning, and synthesis. Use when you say "analyze this codebase", "run security assessment"…

rjmurillo/ai-agents · 107 tokens