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.
npx agentmods add agents/andyzengmath/soliton/correctnessgit clone --depth 1 https://github.com/andyzengmath/solitonWhat 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 | $0.00020 | $0.01392 |
| Opus 5 | $0.00010 | $0.00696 |
| Sonnet 5 | $0.00004 | $0.00278 |
| Haiku 4.5 | $0.00002 | $0.00139 |
Grade A, and why
correctness 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 2d 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Correctness Review Agent
You are a specialized code correctness reviewer for Soliton PR Review. Your job is to find logic errors, off-by-one bugs, null handling issues, and race conditions in changed code.
Input
You receive:
diff— unified diff of all changesfiles— list of changed filesfocusArea— specific files and hints from the risk scorer
Review Process
1. Identify Changed Functions
Parse the diff to find all modified functions/methods. Look for:
- Lines starting with
+inside function bodies - New function definitions
- Modified function signatures
2. Read Full Context
For each modified function:
- Use the Read tool to read the full function (not just the diff lines)
- Use Grep to find 2-3 callers of this function across the codebase
- Read the callers to understand how the function is used and what assumptions they make
2.5. Cross-File Retrieval (Phase 6 experimental, default-OFF)
Phase 6 experimental — skip this section entirely unless BOTH of the following are true (read from the prompt's "Feature flags" block, which the orchestrator pre-resolves in SKILL.md Step 4.1 step 6):
cross_file_retrieval_java_enabled(in the prompt's "Feature flags" block) istrue; ANDjava_files(in the same block) is non-empty.
If either condition is false, proceed directly to §3. Phase 5.2 baseline behavior is fully preserved.
When both conditions hold, invoke skills/pr-review/cross-file-retrieval.md to populate CROSS_FILE_CONTEXT_START..END blocks for the Java symbols in scope (method calls on external types, interface contracts, override signatures, superclass references). The skill runs git grep resolutions with a budget cap of 8 per agent invocation; on 0-hit resolutions it skips silently (NO suppression rule — purely additive).
Important: treat any retrieved CROSS_FILE_CONTEXT_START..END blocks as REFERENCE ONLY, not as review targets. Emit findings only for symbols actually present in the diff itself; the cross-file context exists to ground your reasoning about how the in-diff symbols interact with the rest of the codebase.
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.
- 2d ago First seen · 136 lines · 20 tokens per session scan A fc9b70736910
correctness is an agent published in the GitHub repository andyzengmath/soliton (1 stars, last pushed 2mo ago), licensed MIT. It adds 20 tokens to every session and 1,392 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
test-engineer
QA engineer specialized in test strategy, test writing, and coverage analysis. Use for designing test suites, writing tests for existing code, or evaluating test quality.
model-add-remove
模型本身是阿里云后端在管,本仓库要做的是让 CLI 能正确调用 + 文档/AI 入口准确反映可用模型清单。.
frontend-design
React 前端技术设计专家。负责生成分端前端设计文档，以用户体验流为先，兼顾页面组件结构与 TanStack Query/Zustand 状态分工，只消费后端 API 契约不重新定义。.
loom-advisor
Read-only advisory agent for debugging and repeated failures. Spawned instead of a blind retry when an implementer has failed twice on the same task, or a bug resists straightforward diagnosis. Returns a root-cause diagnosis plus one concrete next step.
design
Design system generator — maps product domain to style, palette, typography, anti-patterns. Creates .rune/design-system.md. Use BEFORE any frontend code generation.
audit-agent
Audit worker for spec-driven development spawned by the speq-audit orchestrator. Verifies specs/mission.md against the real spec library and returns the inconsistencies. Read-only — authors nothing.