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 skills/explyt/spring-plugin/plugin-code-review-orchestratednpx skills add explyt/spring-plugin --skill plugin-code-review-orchestratedgit clone --depth 1 https://github.com/explyt/spring-pluginWhat 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 | $0.00072 | $0.05098 |
| Opus 5 | $0.00036 | $0.02549 |
| Sonnet 5 | $0.00014 | $0.01020 |
| Haiku 4.5 | $0.00007 | $0.00510 |
Grade A, and why
plugin-code-review-orchestrated 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 — 387 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orchestrated code review
You are the review orchestrator. You do not replace the reviewer subskills. Your job:
- normalize the review target;
- build a shared review packet;
- select the relevant reviewer subskills;
- run them in parallel;
- hand the results to validators;
- assemble a single final report.
You do not fix code within this skill. You organize the review, persist artifacts, and show the user the outcome.
Default loop
clarify minimal -> scope resolve -> optional context -> review packet -> routing -> review wave -> per-artifact validation wave -> aggregate final report
Mandatory agent/skill passes
Always run these pairs:
call_ask_agent+ skillreview-scope-resolvercall_review_agent+ skillreview-core-correctnesscall_code_agent+ skillreview-findings-validatorfor each reviewer artifact from the review wave
Optional agent/skill passes (by routing decision)
call_ask_agent+ skillreview-context-buildercall_review_agent+ skillreview-async-lifecyclecall_review_agent+ skillreview-psi-vfs-indexingcall_review_agent+ skillreview-ui-platformcall_review_agent+ skillreview-persistence-security-resourcescall_review_agent+ skillreview-statistic-coveragecall_review_agent+ skillreview-ui-leak-via-listeners
Every subagent call in every wave must state its expected result shape: a brief summary in chat plus the exact path of the artifact file written.
1. Minimal clarification
First determine what exactly must be reviewed.
If the user already gave a technical reference, use it as the input for the call_ask_agent handoff with skill review-scope-resolver: PR, diff, commit, branch, issue, URL, file set, directory or module, or a specific file/class/function.
Ask the user at most one short question, and only if:
- the review target is not defined at all;
- or a blocking ambiguity remains after scope resolution.
Do not ask the user to retell context that can be obtained by investigation. Do not treat currently open files as an implicit review target without explicit confirmation.
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 · 387 lines · 72 tokens per session scan A 980350556bdf
plugin-code-review-orchestrated is a skill published in the GitHub repository explyt/spring-plugin (160 stars, last pushed 4d ago), licensed Apache-2.0. It adds 72 tokens to every session and 5,098 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 skills, from other repositories
analyze-external-methods
Analyze an OpenTaint scan's dropped external methods and decide which of them are propagators and optionally sinks. Use when a dropped-external-methods.yaml needs classification for dropped method type.
create-test-project
Create an OpenTaint test project with positive/negative samples for verifying a rule or approximation. Use when a rule or approximation needs a test project to check against.
appsec-agent
Run an end-to-end OpenTaint application-security analysis while owning the long project build and scans and delegating each other pipeline stage. Use when the user asks to find vulnerabilities, or scan an application for security issues.
create-pass-through-approximation
Model a method's taint propagation as a passThrough approximation. Use for a dropped method whose propagation is simple copying.
discover-attack-surface
Classify project-used dependency members and record the taint sources among them not covered by the built-in rules. Use for the source-discovery depth pass.
analyze-findings
Triage OpenTaint findings statically. Use when scan findings need a TP/FP verdict.