agentfs-eval

A three-level evaluation procedure for checking the health and maturity of an AgentFS workspace. AgentFS is a workspace layout that stores an agent project's files, memory, and related configuration under an .agents directory.

In plain words
What is it for?
Use it when explicitly evaluating a project or lightweight AgentFS workspace, optionally against a directory other than the current one.
Why use it?
It provides progressively deeper checks, including semantic review, so problems can be found without relying only on a quick file inspection.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/rhtevan/agentfs/agentfs-eval
Any agent
npx skills add rhtevan/agentfs --skill agentfs-eval
Clone the repo
git clone --depth 1 https://github.com/rhtevan/agentfs

Made for: Claude Code, Codex.

Per session 16 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,894 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00016 $0.02894
Opus 5 $0.00008 $0.01447
Sonnet 5 $0.00003 $0.00579
Haiku 4.5 $0.00002 $0.00289

Measured 2d ago against content hash 3c150d865b76, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agentfs-eval 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.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/agentfs-behavior.sh, scripts/agentfs-check.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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/agentfs-eval/SKILL.md · 271 lines

How it starts

The opening of the file, as written. The whole thing — 271 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AgentFS Eval

Assess the health and maturity of an AgentFS workspace through three layers of progressively deeper verification.

Overview

Property Value
Version 1.0
Trigger Explicit only — user asks to run eval
Scope Evaluates a PROJECT or LITE-scoped .agents/ directory
LITE scope Skips skills/profiles L1 checks; adds LITE exclusion assertions
Dependencies bash, find, grep, stat, git (optional but recommended)
LLM required Layer 3 only (semantic rubrics)

For the most reliable evaluation, run this skill in a fresh session against the target project directory. This eliminates conversational bias from prior work in the same session.

Example prompts:

"Run agentfs eval" "Run agentfs eval against /home/user/projects/my-project"

The skill works against any directory containing .agents/ — it does NOT need to be the current working directory. If no path is provided, it defaults to the current working directory.

Why a Fresh Session?

  • Eliminates self-evaluation bias — the agent that did the work should not be the one evaluating it
  • Removes conversational context — no memory of what the user "wanted," just cold assessment of filesystem state
  • Reduces sycophancy pressure — no incentive to report good results to please the user who just watched it work
  • More capable models give stronger classification accuracy for Layer 3 semantic rubrics

Maturity Levels

Level Name Requirements
L0 Absent No .agents/ directory exists
L1 Scaffolded .agents/ exists with valid structure
L2 Structurally Sound All Layer 1 assertions pass
L3 Behaviorally Safe Layer 1 + Layer 2 assertions pass
L4 Semantically Accurate Layer 1 + Layer 2 + Layer 3 assertions pass
L5 Self-Correcting Agent can run eval, detect violations, and fix them

Read the full file on GitHub · 271 lines

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. 2d ago First seen · 271 lines · 16 tokens per session scan A 3c150d865b76

Subscribe to this mod's changes

agentfs-eval is a skill published in the GitHub repository rhtevan/agentfs (2 stars, last pushed 3d ago), licensed Apache-2.0. It adds 16 tokens to every session and 2,894 once invoked, about $0.0001 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

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

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.

obra/superpowers · 37 tokens

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.

microsoft/vscode · 53 tokens

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…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

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.

microsoft/vscode · 62 tokens