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/kkollsga/codingest/codingest-code-reviewnpx skills add kkollsga/codingest --skill codingest-code-reviewgit clone --depth 1 https://github.com/kkollsga/codingestWhat 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 | $0.00066 | $0.00990 |
| Opus 5 | $0.00033 | $0.00495 |
| Sonnet 5 | $0.00013 | $0.00198 |
| Haiku 4.5 | $0.00007 | $0.00099 |
Grade A, and why
codingest-code-review 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 yesterday.
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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codingest code review
Use Codingest for structural evidence during review. The graph complements the git diff, source reading, and literal-text search; it does not replace them.
Review workflow
Prerequisites: the
codingestandkgliteCLIs must be on PATH.pip install codingestprovides both plus the builder-awarecodingest-mcpserver. Rust-only environments can alternatively use Cargo.
-
Inspect the diff and repository guidance first. Identify changed symbols and the base/head revisions.
-
Build or refresh the graph without executing repository code:
codingest build . --output .kglite/code-review.kgl --format jsonFor a committed comparison, use one graph spanning both revisions:
codingest build . --revs '<base>' '<head>' \ --output .kglite/code-review.kgl --format json -
Always discover the actual schema before writing Cypher:
kglite describe .kglite/code-review.kgl --connections --cypher -
Query the smallest structural question that can confirm or reject a review hypothesis. Use JSON for agent parsing:
kglite query .kglite/code-review.kgl '<cypher>' --format json -
Open every implicated file and verify the behavior at exact lines. Report only findings supported by source evidence. Do not infer runtime behavior from an edge alone.
-
Before reusing an artifact, check freshness:
codingest status --output .kglite/code-review.kgl --format json
See queries.md for query patterns, public-repositories.md for safe public-repo review, and mcp-upgrade.md for the persistent MCP workflow.
What counts as a finding
The workflow above verifies a finding against exact source lines. This is the prior question — what is eligible to be a finding at all.
- A finding names a concrete failure: the input, state, or sequence, and the wrong outcome it produces. A wrong result, a crash, data loss or corruption, a broken contract with a caller or a persisted format, a security hole, a measured performance regression, a check that cannot fail, or a claim the code contradicts. "No findings" is a valid review, and a good one.
- Design, structure, naming, "consider using X", and "this won't scale" are not findings — they are mis-staged. Their venue is planning, where "I would have designed this differently" is invited and settled before the code exists. After a plan is approved, review measures the implementation against that plan and against correctness, never against a design the reviewer would have preferred. A design opinion formed while reading a diff is input to the next plan.
- A finding that cannot state its failure case is removed, not downgraded. Severity labels are how a preference gets laundered into a report: "Minor: consider extracting this" is a preference wearing a label.
- One narrow exception: citing a rule the project declared before the diff existed — a documented ceiling, a boundary rule, a checklist — naming both the rule and the violating line. That is enforcement, not taste.
- A review tool's effort or confidence level is orthogonal. A higher level buys more speculative bugs; it never buys permission to report preferences.
- A graph edge showing coupling is a fact, not a defect. Structural evidence answers "what would this change reach"; it does not by itself establish that anything is wrong.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 96 lines · 66 tokens per session scan A 36ba11515f63
codingest-code-review is a skill published in the GitHub repository kkollsga/codingest (0 stars, last pushed yesterday), licensed MIT. It adds 66 tokens to every session and 990 once invoked, about $0.0003 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…