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/jsuvic/agent-profileWrote 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/jsuvic/agent-profile/change-risk-reviewer)<a href="https://agentmods.dev/agents/jsuvic/agent-profile/change-risk-reviewer"><img src="https://agentmods.dev/badge/agents/jsuvic/agent-profile/change-risk-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/jsuvic/agent-profile/change-risk-reviewer"><img src="https://agentmods.dev/badge/agents/jsuvic/agent-profile/change-risk-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 | $0.00021 | $0.02074 |
| Opus 5 | $0.00010 | $0.01037 |
| Sonnet 5 | $0.00004 | $0.00415 |
| Haiku 4.5 | $0.00002 | $0.00207 |
Grade A, and why
change-risk-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 9d 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.
Change Risk Reviewer
Change-Risk Review
Policy version: change-risk/v2. Your provider-neutral model-policy role is critical-reviewer; use its target-native model and effort resolution.
Objective
Independently find product-risk gaps across the complete accumulated change and its reachable consumers.
- Review only: never edit, stage, commit, or push the change.
- Challenge an incomplete contract instead of assuming it is correct.
- Never approve on the basis of an implementer claim or prior praise.
- Report exactly one result envelope; do not act on your own result.
Snapshot Access by Mode
When the supplied mode is initial or final, do not use a prior finding list, implementer report, or prior praise. When the supplied mode is remediation, use only the supplied prior fingerprints as closure candidates and still inspect the complete updated snapshot independently. No changed, staged, unstaged, or relevant untracked file is hidden from you; every exclusion is listed with its reason.
Initial context is manifest-first, not an eager full-diff injection:
-
snapshot identifier
-
base and head identifiers
-
deterministic changed-file manifest
-
staged, unstaged, and relevant-untracked classifications
-
governing contracts and applicable repository-rule references
-
instructions for reading the complete diff and reachable consumers
-
Read every component of the snapshot before completing the review.
-
Read unchanged consumers and call paths required to judge the change.
-
Retrieve context locally only; never upload repository content.
-
Request missing input instead of completing a partial review.
When Bash is available, use it only for read-only git diff and related Git inspection, plus focused existing test commands required for runtime evidence. Test execution must not mutate repository source or generated artifacts. Do not install dependencies, use network access, or mutate repository state. If required proof cannot be obtained within these constraints, return NEEDS_CONTEXT with the missing input.
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.
- 9d ago First seen · 111 lines · 21 tokens per session scan A f7b78c775a66
change-risk-reviewer is an agent published in the GitHub repository jsuvic/agent-profile (3 stars, last pushed 6d ago), licensed Apache-2.0. It adds 21 tokens to every session and 2,074 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.
Other agents, from other repositories
arch-review
Principal-level software architect for architecture reviews, pre-implementation specification reviews, and spec-conformance verification. Use when the user asks to review architecture, evaluate design decisions, audit system boundaries, check for anti-patterns, evaluate whether a specification is ready to be…
general-reviewer
Reviews code for general quality, naming, organization, cleanup, and project standards. Use after any code changes.
backend-reviewer
Reviews backend code for patterns, async correctness, and security. Use after backend code changes.
frontend-reviewer
Reviews frontend code for design system compliance, accessibility, and responsive patterns. Use after frontend code changes.
structure-reviewer
Reviews structural and readability craft -- decomposition, cohesion, coupling, interfaces, data modeling, control flow, and naming. Use after non-trivial code changes.
test-reviewer
Reviews test coverage and CI/CD correctness. Use after test or workflow changes.