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 skills add hoatv2211/GameStudio-CodexKIT --skill evidence-first-debugginggit clone --depth 1 https://github.com/hoatv2211/GameStudio-CodexKITWrote 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/skills/hoatv2211/gamestudio-codexkit/evidence-first-debugging)<a href="https://agentmods.dev/skills/hoatv2211/gamestudio-codexkit/evidence-first-debugging"><img src="https://agentmods.dev/badge/skills/hoatv2211/gamestudio-codexkit/evidence-first-debugging/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/skills/hoatv2211/gamestudio-codexkit/evidence-first-debugging"><img src="https://agentmods.dev/badge/skills/hoatv2211/gamestudio-codexkit/evidence-first-debugging.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 | $0.00063 | $0.00953 |
| Opus 5.5 | $0.00025 | $0.00381 |
| Sonnet 5.5 | $0.00013 | $0.00191 |
| Haiku 4.5 | $0.00006 | $0.00095 |
Grade A, and why
evidence-first-debugging 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 1mo 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Evidence-First Debugging
Overview
Find the root cause before changing code. A fix is complete only when the original symptom is reproduced, isolated, repaired minimally, and guarded by regression evidence.
When to use
Use for crashes, wrong behavior, failing tests, startup problems, packet mismatches, offline failures, build errors, or unexpected generated output.
When NOT to use
Do not use for open-ended feature design, routine build verification, broad code review, or mutation without a reproducible symptom.
Required inputs and context discovery
Collect exact symptom, expected behavior, reproduction steps, frequency, environment, first failing version or commit when known, logs, relevant ownership boundaries, and safe instrumentation options.
Safety and risk level
Inspection and instrumentation are preferred. Any mutation follows test-first discipline and exact file ownership; database, services, assets, or external projects need separate safety approval.
Workflow
- Reproduce the symptom with the smallest deterministic command or fixture. Completion criterion: failure occurs for the expected reason, or reproduction is BLOCKED with evidence.
- Trace the data and control path from observed failure toward its source using native inspection and, when available, a fresh code-intelligence provider. Completion criterion: component boundaries and the first incorrect state are identified; inferred, dynamic, generated, and stale graph edges are labeled rather than treated as causal proof. Source, logs, tests, and runtime evidence own the root-cause verdict.
- Rank hypotheses and add minimal instrumentation that distinguishes them. Completion criterion: one hypothesis is supported and alternatives are weakened by output.
- Write a failing regression test or executable check before the fix. Completion criterion: the check fails on the original behavior.
- Apply the smallest root-cause fix and rerun the focused check. Completion criterion: the focused check passes without weakening assertions.
- Run adjacent and broader verification, then record limitations. Completion criterion: regressions are checked or explicitly BLOCKED.
What ships with it
1 file 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.
- 1mo ago Changed · +1 lines 8d96b86d5ad1
- 1mo ago First seen · 85 lines · 63 tokens per session scan A 76ce669c7dca
evidence-first-debugging is a skill published in the GitHub repository hoatv2211/GameStudio-CodexKIT (3 stars, last pushed 1mo ago), licensed MIT. It adds 63 tokens to every session and 953 once invoked, about $0.0003 per session on Opus 5.5. 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
behavior-contract
Bug condition/postcondition formalization as testable Behavior Contracts. Defines invariants that must be preserved across fixes.
quality-hooks
Language-specific auto-lint/format/typecheck pipeline. Supports Python (ruff+pyright), TypeScript (prettier+eslint+tsc), Go (gofmt+golangci-lint). Auto-fix and convergence loops.
debug
Run /debug to find and fix a bug's root cause: a test failing for an unclear reason, /check verify finding a failure, or behavior being wrong. Runs a reproduce, localize, hypothesize, test, fix, verify loop, makes the minimal fix, and hands a regression test to /test. No features, no extra refactors.
prove-checks
Prove a passing check was capable of failing before recording it as evidence. Use when a test, CI job, build-and-diff, smoke test or rehearsal comes back green and that green is about to be treated as proof - especially when the check depends on a setup mutation (a sed/awk rewrite, an env var, a secret, a fixture…
team-fix
Use for the full bug-fix pipeline only on explicit request. Never infer from an ordinary bug fix. Runs the compressed pipeline.
reliability
Use when user asks to stress-test the system, validate reconnection behavior, check for race conditions, or verify end-to-end data flow integrity.