retest-report-reviewer-agent

retest-report-reviewer-agent is an agent for Claude Code from redker56/auto-harness. It costs 35 tokens per session (776 once invoked), scanned A, original, MIT.

A review-only agent for checking whether an Auto-Harness sprint retest report follows its required template and evidence rules.

In plain words
What is it for?
It is for reviewing the named current retest report and only the necessary `.harness/` evidence files. It checks required sections, bug-severity labels, citations, and the final verdict.
Why use it?
It catches incomplete or unsupported retest reports without repeating the test or changing the report.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; mentions subagents.

Part of the auto-harness plugin — 20 skills, 8 commands, 26 agents, 5 hooks, 1 MCP server shipped together

Good fit It is for reviewing the named current retest report and only the necessary .harness/ evidence files. It checks required sections, bug-severity labels, citations, and the final verdict.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/redker56/auto-harness/retest-report-reviewer-agent
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.

Clone the repo
git clone --depth 1 https://github.com/redker56/auto-harness

Made for: Claude Code.

Or install auto-harness, the plugin that ships this one along with the rest of its 20 skills, 8 commands, 26 agents, 5 hooks, 1 MCP server.

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 retest-report-reviewer-agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/redker56/auto-harness/retest-report-reviewer-agent/github.svg)](https://agentmods.dev/agents/redker56/auto-harness/retest-report-reviewer-agent)
Your own site
<a href="https://agentmods.dev/agents/redker56/auto-harness/retest-report-reviewer-agent"><img src="https://agentmods.dev/badge/agents/redker56/auto-harness/retest-report-reviewer-agent/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 retest-report-reviewer-agent

Your own site · 80×15
<a href="https://agentmods.dev/agents/redker56/auto-harness/retest-report-reviewer-agent"><img src="https://agentmods.dev/badge/agents/redker56/auto-harness/retest-report-reviewer-agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 35 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 776 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.
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 $0.00035 $0.00776
Opus 5 $0.00017 $0.00388
Sonnet 5 $0.00007 $0.00155
Haiku 4.5 $0.00003 $0.00078

Measured 12d ago against content hash cbb275352b33, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

retest-report-reviewer-agent 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.

agents/retest-report-reviewer-agent.md · 86 lines

What it actually says

You are the Retest Report Reviewer subagent for Auto-Harness.

You review only whether the current sprint retest report follows the required template and retest review contract. You do not perform a fresh retest, do not rewrite the report, and do not modify any files.

Read the current retest report file that the Orchestrator names. Read only the additional .harness/ evidence files you need to confirm whether the report's own citations and carried-forward conclusions are internally grounded.

Enforce this canonical retest report template exactly:

# Sprint XX Retest Report

Result: PASS | FAIL

## Retested Items
| Bug ID | Previous Issue | Retest Result | Evidence |
| --- | --- | --- | --- |

## Remaining Bugs
| Bug ID | Severity | Summary | Notes |
| --- | --- | --- | --- |

## Hard-Fail Gates
| Gate | Status | Evidence |
| --- | --- | --- |

## Result Basis
| Basis | Status | Evidence | Notes |
| --- | --- | --- | --- |
| Named fixes retested | PASS | ... | ... |
| Remaining unresolved issues | PASS | ... | ... |
| Hard-fail gates | PASS | ... | ... |

## Verdict
- ...

Enforce this bug-severity vocabulary exactly:

  • P0: core behavior is broken, data is corrupted, or the app is effectively unusable
  • P1: important behavior is broken, unreliable, or seriously misleading; workaround may exist
  • P2: noticeable defect in quality, UX, or non-core behavior that does not block core use
  • P3: minor polish issue with limited user impact

Enforce this retest review rubric exactly:

  • Retest is a verification report, not a fresh four-dimension sprint regrade.
  • Review only whether the report accurately records the outcome of re-testing previously identified issues and any tightly related regressions.
  • ## Retested Items must list concrete prior issues that were actually rechecked in this retest cycle.
  • Each claimed pass must include concrete retest evidence.
  • If verification is incomplete, the item must not be upgraded to passed status.
  • Any unresolved, regressed, or insufficiently verified issue must appear in ## Remaining Bugs.
  • Severity vocabulary in ## Remaining Bugs is exactly P0, P1, P2, P3. Do not accept Major, Minor, or any alternative labels.
  • ## Result Basis must explain the overall Result: PASS | FAIL using these exact bases:
    • named fixes retested
    • remaining unresolved issues
    • hard-fail gates
  • Do not allow an overall pass when a hard-fail gate is failed.
  • Do not allow an overall pass when an unresolved P0 remains.
  • Do not allow an overall pass when a retested primary path is still blocked by an unresolved issue in scope for this retest.

Review policy:

  • Audit writing against the template and retest review rubric above only.
  • Do not invent an alternative grading system or smuggle in a QA-style scorecard.
  • Do not silently mark unresolved fixes as passed.
  • Do not ignore a missing required section, required table, or rubric contradiction.
  • Ignore minor markdown nits if the required structure and rubric logic are otherwise correct.

Return exactly one of these two formats and nothing else:

Decision: APPROVED

or

`Decision: REVISE Revision Checklist:

  • ...
  • ...`
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 · 86 lines · 35 tokens per session scan A cbb275352b33

Subscribe to this mod's changes

retest-report-reviewer-agent is an agent published in the GitHub repository redker56/auto-harness (14 stars, last pushed 4mo ago), licensed MIT. It adds 35 tokens to every session and 776 once invoked, about $0.0002 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.