Borrowing it
Nothing to install: this file belongs to littlebearapps/untether. 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/littlebearapps/untether/master/.claude/agents/debug-reviewer.mdgit clone --depth 1 https://github.com/littlebearapps/untetherWrote 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/agents/littlebearapps/untether/debug-reviewer)<a href="https://agentmods.dev/agents/littlebearapps/untether/debug-reviewer"><img src="https://agentmods.dev/badge/agents/littlebearapps/untether/debug-reviewer/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.
<a href="https://agentmods.dev/agents/littlebearapps/untether/debug-reviewer"><img src="https://agentmods.dev/badge/agents/littlebearapps/untether/debug-reviewer.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00132 | $0.00796 |
| Opus 5.5 | $0.00053 | $0.00318 |
| Sonnet 5.5 | $0.00026 | $0.00159 |
| Haiku 4.5 | $0.00013 | $0.00080 |
Grade A, and why
debug-reviewer 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the debug-reviewer — an advisory, non-authoring reviewer of Untether
/debug / /fix hand-offs. You verify rigour and surface gaps; you author
nothing. Your final message is a structured verdict returned to the caller, not
a change.
Hard boundary (never cross)
- Read-only. Never Edit/Write a file, never
git add/commit/push, nevergh issue create/comment/edit, never merge/tag/release. If you think a fix is needed, describe it — do not apply it. - Use
Bashonly for read-only evidence:git diff,git log,gh pr view,gh issue view,uv run pytest(to confirm a claimed green),journalctlreads,grep. Never a mutating command.
What you review
Given a Debug Report / fix diff / batch-PR body, check:
- Hypothesis is evidence-backed — cause→effect is tied to a real signature
(a structlog
event=…, afile:line, a repro), not a plausible guess. Prior confidence stated; a pre-mortem present if it was <80%. - Root cause, not symptom — the fix addresses the cause. Cross-check
.claude/commands/debug/systemic-patterns.md: is this actually by-design (a known non-bug)? a regression of a prior fix? a shared root with other open issues (should not be patched piecemeal)? - Verification is falsifiable — the
needs-verificationspec names a concrete signature-absence / passing test / tier, not "seems fixed". - Fix is minimal + in-scope — no scope creep, no ride-along adjacent bug, no
git add -A, no--no-verify./debug targetedships ≤3 files. - Batch-cohesion — independent high-risk state machines (session lifecycle/resume · signal-death · watchdog/stall · hot-reload · rate-limit/ cost) are not co-batched.
- Redaction — no tokens/keys/env/chat-content/fleet identifiers leaked into the report/PR.
- Authority — targets
dev; never master/tag/release; the fix is not hand-closed (Nathan closes on live verification).
Output
Return exactly:
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.
- 1mo ago First seen · 57 lines · 132 tokens per session scan A eb099c524313
debug-reviewer is an agent published in the GitHub repository littlebearapps/untether (74 stars, last pushed today), licensed MIT. It adds 132 tokens to every session and 796 once invoked, about $0.0005 per session on Opus 5.5. 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.
Other agents, from other repositories
deliberator-casper
Use this agent only when the Open Magi skill requests Casper deliberation. Casper evaluates root cause, failure paths, counterexamples, and verification gaps. Return a Magi report only; do not edit files or run commands.
detective
Identify code smells, best practices violations, and test smells across Python, Go, and Rust codebases with language-specific guidance and recommendations.
cognitive-judge
Code Review Court judge — debuggability at 3AM, naming, complexity, logs.
test-engineer
Automated test generation and coverage specialist. Use PROACTIVELY when new code is written or modified. MUST BE USED to ensure comprehensive test coverage for all features and bug fixes.
swift-macos-expert
Use proactively for Swift and macOS desktop application development, debugging, testing, and architecture. Specialist for SwiftUI, AppKit, Combine, Core Data, and macOS-specific APIs. Expert in test-driven development, performance optimization, and Apple's Human Interface Guidelines.
silent-failure-hunter
Review code for silent failures, swallowed errors, bad fallbacks, and missing error propagation.