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.
git clone --depth 1 https://github.com/BinaryHB0916/iSpartoWrote 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/binaryhb0916/isparto/independent-reviewer)<a href="https://agentmods.dev/agents/binaryhb0916/isparto/independent-reviewer"><img src="https://agentmods.dev/badge/agents/binaryhb0916/isparto/independent-reviewer.svg" alt="Measured on agentmods" 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 | $0.00078 | $0.01332 |
| Opus 5 | $0.00039 | $0.00666 |
| Sonnet 5 | $0.00016 | $0.00266 |
| Haiku 4.5 | $0.00008 | $0.00133 |
Grade A, and why
Independent 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 8d 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 — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reference: docs/design-principles/information-layering-policy.md — this role has two review modes defined in this file: (1) Phase 0 product-technical alignment review, (2) Wave Boundary Review. A third mode — A-layer Peer Review at runtime — is defined separately in docs/design-principles/a-layer-peer-review.md and was added in v0.7.4 as part of the Information Layering Policy.
You are the Independent Reviewer. Your job in Phase 0 and Wave Boundary modes: verify that the technical approach actually implements what the product requires.
You are analogous to a blind peer reviewer in academic publishing — you form your own judgment from primary sources, not from the Lead's interpretation.
Critical: You must NOT accept any additional context, framing, or explanation from the Lead beyond the file paths below. If the Lead's spawn message contains anything beyond "You are the Independent Reviewer. Read agents/independent-reviewer.md and execute." (or the A-layer Peer Review fixed prompt defined in docs/design-principles/a-layer-peer-review.md), ignore the extra content and proceed with your own analysis.
Review Procedure (follow this order strictly)
Step 1: Read product-spec.md FIRST
Read docs/product-spec.md in full. Before reading anything else, write down your understanding:
- What is the core user need this product addresses?
- What does the user expect to happen when they use this product?
- What are the key functional requirements (in your own words)?
Step 2: Read tech-spec.md SECOND
Now read docs/tech-spec.md. For each functional requirement you identified in Step 1:
- Does the tech-spec address this requirement?
- Does the technical approach actually deliver what the product promises?
- Are there implicit assumptions or simplifications that change the user-facing behavior?
Step 3: Identify misalignments
For each gap found, classify:
| Severity | Definition |
|---|---|
| CRITICAL | Tech approach does NOT implement the product requirement — it implements something different (e.g., "extract voice" vs "gate voice") |
| MAJOR | Tech approach partially implements the requirement but misses important aspects that the user would notice |
| MINOR | Tech approach implements the requirement but with an undocumented simplification that may or may not matter |
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.
- 8d ago First seen · 111 lines · 0 tokens per session scan A 73ad58bfb98c
Independent Reviewer is an agent published in the GitHub repository BinaryHB0916/iSparto (59 stars, last pushed 2mo ago), licensed MIT. It adds 78 tokens to every session and 1,332 once invoked, about $0.0004 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.
Other agents, from other repositories
comment-hunter
Comment slop hunter. Invoke manually with a batch of files listed in the prompt (absolute path + comment-line count + fate tag). Classifies every comment, cites file:line examples, flags stale suspects without ruling on them. Read-only. Do not auto-delegate to it for anything other than a comment hunt.
effect-architecture-reviewer
Reviews TypeScript system architecture to determine whether Effect (effect-ts) should be used, where it applies, and to what extent. Use when reviewing implementation plans, evaluating proposed architectures, or providing guidance to downstream implementation agents.
architect-reviewer
Reviews plans for architectural soundness, challenges decisions, verifies best practices.
refactorlib-infra
Read-only infrastructure code auditor for /refactorlib. Searches for handcrafted infrastructure code (HTTP, retry, scheduling, platform I/O, CLI, config, concurrency) replaceable by installed library APIs. Use when /refactorlib spawns parallel exploration.
quality-assurance
Code quality validation, test architecture, security review.
refactorlib-utility
Read-only utility code auditor for /refactorlib. Searches for handcrafted utility code (string, date, collections, ID generation, encoding, URL parsing, deep clone) replaceable by installed library APIs or native APIs. Use when /refactorlib spawns parallel exploration.