Code Review Walkback

Code Review Walkback is an agent for Claude Code from Peter-N91/hve-squad-mcp. It costs 29 tokens per session (816 once invoked), scanned A, original, MIT.

A small review agent that performs bounded deep investigations for a code-review process and attaches the evidence to the originating review-board item.

In plain words
What is it for?
Use it when a code review needs evidence-based investigation of a specific question, with results stored as a Register 2 artifact linked to the relevant board item.
Why use it?
It turns focused research into a standard review record without inventing review categories or duplicating the broader research process.

Agent for Claude Code

Written for Claude Code: user-invocable in frontmatter. Also seen: mentions subagents.

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.

agentmods
npx agentmods add agents/peter-n91/hve-squad-mcp/code-review-walkback
Clone the repo
git clone --depth 1 https://github.com/Peter-N91/hve-squad-mcp

Made for: Claude Code.

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 Code Review Walkback

README.md
[![agentmods](https://agentmods.dev/badge/agents/peter-n91/hve-squad-mcp/code-review-walkback.svg)](https://agentmods.dev/agents/peter-n91/hve-squad-mcp/code-review-walkback)
Your own site
<a href="https://agentmods.dev/agents/peter-n91/hve-squad-mcp/code-review-walkback"><img src="https://agentmods.dev/badge/agents/peter-n91/hve-squad-mcp/code-review-walkback.svg" alt="Measured on agentmods" height="20"></a>
Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 816 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00029 $0.00816
Opus 5 $0.00015 $0.00408
Sonnet 5 $0.00006 $0.00163
Haiku 4.5 $0.00003 $0.00082

Measured 5d ago against content hash 4ab329063bfb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

Code Review Walkback 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 5d 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.

host/cast/.github/agents/code-review-walkback.agent.md · 47 lines

How it starts

The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Review Walkback

Thin walk-back subagent for the Code Review orchestrator. It activates rpi-research for deep investigative questions, then repackages the completed primary research evidence as a Register 2 artifact anchored to the originating board item.

Skill Reference Contract

At the start of the run, locate the skill named code-review and read these references from it (paths are relative to that skill) exactly once in a single parallel read_file block, then apply them verbatim:

  • SKILL.md (skill entrypoint)
  • references/dispatch-loop.md
  • references/output-formats.md

Do not invent severity levels, categories, or output fields the skill does not define.

Lane Preset

  • Perspective: Deep investigation.
  • Register: Register 2.
  • Research boundary: Stay structured and evidence-based. Do not turn this into a generic summary or duplicate the rpi-research workflow.

Required Steps

  1. Read input. Read diff-state.json once for branch, base, files, findingsFolder, boardItem, question, investigationId, researchTopic, researchPurpose, researchAudienceUse, researchQuestions, evidenceCriteria, researchScope, researchNonGoals, researchConstraints, suppliedEvidence, researchRequestedOutputs, researchOutputMode, trustedEvidenceRoot, and register2ArtifactPath. In the same parallel block, read the Skill Reference Contract files.
  2. Validate the activation inputs. Require every scoped input to be explicit. Confirm that trustedEvidenceRoot equals findingsFolder, is explicitly trusted by the parent, and is distinct from diff-state.json. Confirm that register2ArtifactPath is beneath <findingsFolder>/walkback/ and contains no unresolved placeholder. Return Needs clarification without writing when an input or path is missing or invalid.
  3. Activate research. Activate rpi-research with the topic, purpose, audience and intended use, questions, evidence criteria, scope and non-goals, constraints, supplied evidence, requested outputs, output mode, and trusted alternate evidence root. Require the skill to mirror research/YYYY-MM-DD/<task-slug>-research.md and research/subagents/... beneath that root. Let the skill resolve the exact date, task slug, primary and delegated artifact paths, worker selection, lane contracts, budgets, and research synthesis.
  4. Anchor the result. When research completes, read the returned primary research artifact once, then create or update the Register 2 artifact at register2ArtifactPath. Include the board item id, research question, evidence summary, references, unresolved evidence, and follow-on questions. Preserve links and selectable symbols for later board merge. Treat Blocked or Needs clarification as unresolved evidence: record only the status and smallest blocker when the Register 2 path is valid, then stop.
  5. Return a concise summary. Return the Register 2 artifact path, primary research artifact path, execution status, and a short board-item status note. Do not repeat the primary artifact in the response.

Read the full file on GitHub · 47 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. 5d ago First seen · 47 lines · 29 tokens per session scan A 4ab329063bfb

Subscribe to this mod's changes

Code Review Walkback is an agent published in the GitHub repository Peter-N91/hve-squad-mcp (0 stars, last pushed 2d ago), licensed MIT. It adds 29 tokens to every session and 816 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 2026-08-31.

Related

Other agents, from other repositories

cpp-reviewer

Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.

affaan-m/ECC · 41 tokens

reviewer

Read-only reviewer for an SDD implementation — checks that the change satisfies the acceptance criteria it claims (stage 1) and meets quality/convention/edge-case bars (stage 2). Use after a task (or the whole feature) reaches GREEN, before it's considered done. It reads the diff and the upstream artifacts and reports…

genkovich/sdd · 81 tokens

atomic-auditor

Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…

damusix/atomic-claude · 169 tokens

bt6-pr-auditor

Reviews one pull request in a BT6 codebase for correctness, research integrity, security, verification quality, and merge readiness.

elder-plinius/T3MP3ST · 32 tokens

Reviewer

Mandatory fast reviewer: validates every agent delegation output before acceptance. Checks acceptance criteria, file partitions, regressions, type safety, security basics.

monkilabs/opencastle · 30 tokens

security-auditor

Use this agent when reviewing local code changes or pull requests to identify security vulnerabilities and risks. This agent should be invoked proactively after completing security-sensitive changes or before merging any PR.

NeoLabHQ/context-engineering-kit · 40 tokens