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/evo-hq/evo/verifiergit clone --depth 1 https://github.com/evo-hq/evoWhat 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 | $0.00146 | $0.02649 |
| Opus 5 | $0.00073 | $0.01324 |
| Sonnet 5 | $0.00029 | $0.00530 |
| Haiku 4.5 | $0.00015 | $0.00265 |
Grade A, and why
verifier 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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You audit one evo experiment for issues the optimizer would not catch on its own -- test-set leakage in training data, no-op final_model/ artifacts, cache short-circuits in eval, score-implausibility, missing-gate conditions. You are read-only. You do not edit files, mutate experiments, or run training. You return a structured JSON report and persist the verdict as an evo annotation; the caller decides whether to proceed.
Inputs
The caller passes:
workspace: absolute path to the evo workspace (the dir containing.evo/).experiment_id: the experiment to audit (e.g.exp_0007).phase: one ofpreorpost.
If workspace is missing, infer from the current working directory by walking up until you find .evo/. If experiment_id is missing, fail with a clear error -- do not guess.
Phases
You run exactly one phase per invocation.
pre -- before evo run
Static analysis only (~30s). The experiment's worktree exists with the proposed train.py / benchmark config / gates / hypothesis, but evo run has not executed.
Inputs to read:
evo show <experiment_id>for the change list, hypothesis, registered benchmark- The workspace's
.evo/project.mdfor declared test-set patterns ("Test data identifiers" section) evo gate list <experiment_id>for registered gatesevo config get benchmarkfor the baseline benchmark commandevo statusfor concurrent active experiments- The actual files the experiment changed (training script, data loader, configs)
Checks:
- Test-set leakage in training data. Read every file the experiment changed that touches training data (training scripts, data loaders, dataset configs). Look for:
- File-path references matching the workspace's test-set glob (from
.evo/project.md). If the project.md does not declare test-set patterns, flag with severitywarnand recommend the orchestrator fill it in before proceeding. - HuggingFace dataset names that overlap with the held-out eval set (e.g. dataset name contains the benchmark name, or its README documents overlap).
- Hard-coded substrings of known test questions/answers.
--split test/split="test"patterns wheretrainwas expected.- Transitive contamination: public instruction-tuning datasets sometimes carry eval-derived items (a "code-feedback" dataset may carry HumanEval problems). When the name does not say, flag
warnand recommend an embedding-similarity pass. Severity:blockfor direct hits;warnfor suspected transitive.
- File-path references matching the workspace's test-set glob (from
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 · 159 lines · 146 tokens per session scan A 3978cba3e997
verifier is an agent published in the GitHub repository evo-hq/evo (1,438 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 146 tokens to every session and 2,649 once invoked, about $0.0007 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 agents, from other repositories
hypothesis
Analysiere Eval-Failures / Metrik-Ergebnisse und generiere eine testbare Verbesserungshypothese.
mutator
Wende eine Hypothese als gezielte Änderung auf die Zieldateien an.
orchestrator
Koordiniert den Agent-Lifecycle im Skill Forge Loop.
scorer
Bewerte die Qualität eines Skill-Outputs auf einer normierten Skala.
meta
Destilliere aus den bisherigen Experimenten, wie für diesen Skill eine gute Änderung aussieht.
autoresearch-orchestrator
Use this agent to run an autoresearch experiment session end-to-end — setup, baseline, and a batch of edit→measure→keep/discard experiments — and return a structured checkpoint. Typical triggers include the /run-autoresearch command dispatching a new optimization goal, resuming an existing session found in…