GitEasy: Skill for Claude Code

.claude/skills/honest-completion/SKILL.md

honest-completion is a skill for Claude Code from greenmtnsun/GitEasy. It costs 140 tokens per session (1,943 once invoked), scanned A, original, MPL-2.0.

A completion-checking rule for coding work. It requires claims such as “fixed,” “tested,” or “committed” to be backed by a tool result, and requires unfinished code to be marked as unfinished.

In plain words
What is it for?
Use it whenever an agent is about to claim that work is done, tested, merged, or committed. It also applies when checking work produced by sub-agents.
Why use it?
It prevents partial work or unverified results from being reported as complete. That gives the next person a clearer and more accurate view of the project.

Skill for Claude Code

Written for Claude Code: PreToolUse hook event. Also seen: mentions subagents.

This is greenmtnsun/GitEasy's own configuration. It tells Claude Code how to work on GitEasy itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything GitEasy configures →

Reuse

Borrowing it

Nothing to install: this file belongs to greenmtnsun/GitEasy. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/greenmtnsun/GitEasy/main/.claude/skills/honest-completion/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/greenmtnsun/GitEasy

Made for: Claude Code.

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 honest-completion

README.md
[![agentmods](https://agentmods.dev/badge/skills/greenmtnsun/giteasy/honest-completion/github.svg)](https://agentmods.dev/skills/greenmtnsun/giteasy/honest-completion)
Your own site
<a href="https://agentmods.dev/skills/greenmtnsun/giteasy/honest-completion"><img src="https://agentmods.dev/badge/skills/greenmtnsun/giteasy/honest-completion/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 honest-completion

Your own site · 80×15
<a href="https://agentmods.dev/skills/greenmtnsun/giteasy/honest-completion"><img src="https://agentmods.dev/badge/skills/greenmtnsun/giteasy/honest-completion.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 140 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,943 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 $0.00140 $0.01943
Opus 5.5 $0.00056 $0.00777
Sonnet 5.5 $0.00028 $0.00389
Haiku 4.5 $0.00014 $0.00194

Measured today against content hash b0afc57c1dc2, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

honest-completion 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 today.

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.

.claude/skills/honest-completion/SKILL.md · 108 lines

How it starts

The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.

See also: verify-claims-on-disk for the consumer-side discipline (trust no other-agent claim before verifying); software-creation §6 for the four-tier confidence model.

Honest completion

Two failure modes, one rule, because they're the same lie ("I claimed done when it isn't"):

  • Stealth stubs — handing back partial work as if it were complete.
  • Invented verification — claiming verified / tested / works / passes without citing the tool call that proves it.

Keith named the line directly on 2026-05-12: "They tell me they are ready to move on... only built a stub, or 20, 30 percent. When I asked if they have code checked they will and nothing is revealed. I'm tired of it." Two real incidents in the seven days prior (2026-05-06 dispatch claimed 3 fixes that didn't apply; 2026-05-08 overnight scheduled task shipped 0 of 6 fixes) made the cost concrete. This skill treats the rule as an absolute, not a lean-toward.

When this skill applies

This skill activates the moment a Claude turn contains a completion-claim word:

  • done / completed / finished / shipped / landed
  • implemented / fixed / merged / committed

A Stop hook (no-stealth-completion-stop.ps1) enforces this mechanically — it blocks any turn that contains a completion word and lacks both <verified> and <not-verified> markers.

Rule 1 — Stubs are allowed; undisclosed stubs are not

Any function that is incomplete carries a # STUB: block immediately above it. The # prefix is part of the marker so a grep finds these without false positives in string literals.

# STUB:
# 1. Validate $Path is a real directory (Test-Path -PathType Container)
# 2. Return $null instead of throwing when not found (matches Get-Foo convention)
# 3. Pester test: rejects relative paths
# 4. Pester test: handles UNC paths
# 5. Error path: surface ItemNotFoundException with friendly message
function Get-FooBar { throw "not implemented" }

Rules for the block:

  • Each line is one independently testable obligation. Bullet points like "make it work" do not count.
  • Tests and edge cases must appear, not just the happy path.
  • When an item is implemented AND verified, delete the line in the same edit that implements it. Don't let the block lie.
  • A function with no # STUB: block is a contract that the function is complete and verified.

Read the full file on GitHub · 108 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. today First seen · 108 lines · 140 tokens per session scan A b0afc57c1dc2

Subscribe to this mod's changes

honest-completion is a skill published in the GitHub repository greenmtnsun/GitEasy (1 stars, last pushed yesterday), licensed MPL-2.0. It adds 140 tokens to every session and 1,943 once invoked, about $0.0006 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 2026-11-18.

Related

honest-completion compared with skills from other repositories

honest-completion next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
honest-completiongreenmtnsun/GitEasySkill1yesterdayMPL-2.0A140Claude Code
autoreviewopenclaw/clawhubSkill9.6ktodayMITA17Claude Code, Codex
agent-teams-simplify-and-hardenLeoYeAI/openclaw-master-skillsSkill2.2k6d agoMITA115Claude Code, Codex
revdiff-planumputun/revdiffSkill961todayMITA84Claude Code
re0-mergeLilMGenius/paperthinSkill1.5k3d agoMITA51Claude Code, Codex
explore-unknownsdzhng/skillsSkill1.5k5d agoMITA162Claude Code, Codex
asplos-writing-stylebrycewang-stanford/Awesome-Journal-SkillsSkill1.4k1mo agoMITA65Claude Code