agentic-evals

agentic-evals is a skill for Codex from browoz/agentic-sdlc-skills. It costs 95 tokens per session (470 once invoked), scanned A, original, MIT.

A planning skill for deciding how an AI agent feature will be tested and judged. It separates fixed, repeatable checks from tests for behaviours that can vary.

In plain words
What is it for?
Use it to write test plans, evaluation criteria, trajectory checks, quality rubrics, gold sets, and CI approval gates for agentic systems.
Why use it?
It prevents teams from treating an agent as correct without clear success criteria. It also makes unreliable or subjective results easier to evaluate consistently.

Skill for Codex

Written for Codex: agents/openai.yaml present.

Good fit Use it to write test plans, evaluation criteria, trajectory checks, quality rubrics, gold sets, and CI approval gates for agentic systems.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/browoz/agentic-sdlc-skills/agentic-evals
Install

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.

Any agent
npx skills add browoz/agentic-sdlc-skills --skill agentic-evals
Clone the repo
git clone --depth 1 https://github.com/browoz/agentic-sdlc-skills

Made for: Codex.

Wrote 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.

agentmods badge for agentic-evals

README.md
[![agentmods](https://agentmods.dev/badge/skills/browoz/agentic-sdlc-skills/agentic-evals/github.svg)](https://agentmods.dev/skills/browoz/agentic-sdlc-skills/agentic-evals)
Your own site
<a href="https://agentmods.dev/skills/browoz/agentic-sdlc-skills/agentic-evals"><img src="https://agentmods.dev/badge/skills/browoz/agentic-sdlc-skills/agentic-evals/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.

agentmods 80×15 button for agentic-evals

Your own site · 80×15
<a href="https://agentmods.dev/skills/browoz/agentic-sdlc-skills/agentic-evals"><img src="https://agentmods.dev/badge/skills/browoz/agentic-sdlc-skills/agentic-evals.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 95 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 470 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00095 $0.00470
Opus 5 $0.00048 $0.00235
Sonnet 5 $0.00019 $0.00094
Haiku 4.5 $0.00010 $0.00047

Measured 8d ago against content hash 795bec5af549, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

agentic-evals 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 8d 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.

skills/agentic-evals/SKILL.md · 52 lines

What it actually says

Agentic Evals

Use this skill to define how an agentic feature will be judged before treating it as correct. Keep deterministic tests separate from probabilistic evals.

Workflow

  1. Identify the target: SPEC.md, existing code, agent behavior, tool contract, or production workflow.
  2. Split behavior into:
    • Deterministic: pure logic, schemas, permissions, parsing, API contracts.
    • Non-deterministic: planning, tool choice, summaries, ranking, judgment, multi-step recovery.
  3. For deterministic behavior, write Arrange-Act-Assert test cases and expected failure modes.
  4. For non-deterministic behavior, define evals across task completion, correctness, efficiency, and safety.
  5. Use trajectory evals when tool choice, ordering, or recovery path matters.
  6. Use LLM-as-judge only with a task-specific rubric and a small gold set; prefer pairwise comparison when comparing versions.
  7. Generate EVAL_PLAN.md from assets/templates/EVAL_PLAN.md.

Gate Design

  • Level 1: fast PR checks such as schemas, fixtures, tool contract tests.
  • Level 2: deeper nightly or pre-release evals with repeated trials.
  • Level 3: deployment gate using acceptance thresholds and manual review for high-impact changes.

References

Read ../agentic-engineering-sdlc/references/day4_security_and_evaluation.md for deeper evaluation, judge calibration, and trajectory guidance.

Done Criteria

  • Success criteria are measurable.
  • Deterministic tests and non-deterministic evals are separated.
  • Trial count, thresholds, and gold-set source are documented.
  • CI gate ownership and failure response are clear.
Files

What ships with it

2 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.

Changes

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.

  1. 8d ago First seen · 52 lines · 95 tokens per session scan A 795bec5af549

Subscribe to this mod's changes

agentic-evals is a skill published in the GitHub repository browoz/agentic-sdlc-skills (3 stars, last pushed 2mo ago), licensed MIT. It adds 95 tokens to every session and 470 once invoked, about $0.0005 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.

Related

Other skills, from other repositories

create-custom-grader

Use when converting an existing benchmark, rubric, verifier, task YAML/JSON, or domain check into SkillEvaluator BYOG/BYOT custom evaluation.

NVIDIA/SkillEvaluator · 35 tokens

agent-eval

Use when measuring whether an LLM or agent system actually got better and gating merges on it: golden sets, fixing an inflated LLM-as-judge, scoring RAG (faithfulness, contextual recall) or agent trajectories (tool correctness, completion), or picking an eval framework. NOT building the agent loop, tools or RAG…

ericrisco/rsc-harness · 79 tokens

ai-agent-tool-abuse-and-privilege-escalation

Test AI agent systems for tool abuse, unauthorized actions, privilege escalation through tool chaining, and safety bypass via agentic workflows. Use this skill when assessing autonomous AI agents that use tool-calling (function calling, plugins, actions) to interact with external systems. Covers multi-step attack…

ShulkwiSEC/bb-huge · 83 tokens

matlab-testing-ci

MATLAB R2026a testing, validation, and CI workflow for matlab.unittest, MATLAB Test, Simulink Test, coverage, artifact generation, GitHub Actions, and open-source reproducibility. Use whenever MATLAB work needs verification, regression tests, CI, GitHub publication, or acceptance criteria.

wzyn20051216/matlab-agent-skills · 66 tokens

superpowers

Always-on. The main development method: Plan → TDD → Implement → Verify → Report. Use for ANY non-trivial task. Do not write code without a plan and a test. Complex tasks (>3 files) → split into atomic tasks. Bug fix → Prove-It Pattern (reproduce with a test before the fix).

oleg494/coding-kit · 70 tokens

test-driven-development

Drives development with tests. Use when implementing any logic, fixing any bug, or changing any behavior. Use when you need to prove that code works, when a bug report arrives, or when you're about to modify existing functionality.

oleg494/coding-kit · 50 tokens