What is repo-task-proof-loop CLAUDE.md?

A proof workflow for substantial code changes, refactors, and bug fixes. It requires a written task specification, acceptance checks, evidence, and a fresh verification pass before work is considered complete.

Use it to freeze task requirements, implement against numbered acceptance criteria, collect test evidence, fix failed checks, and verify the final result.

It prevents completion claims based only on earlier chat messages or stale test results by checking the current code against explicit requirements.

How DenisSergeevitch/repo-task-proof-loop configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to DenisSergeevitch/repo-task-proof-loop. 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/DenisSergeevitch/repo-task-proof-loop/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/DenisSergeevitch/repo-task-proof-loop

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/denissergeevitch/repo-task-proof-loop/claude-md/github.svg)](https://agentmods.dev/instructions/denissergeevitch/repo-task-proof-loop/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/denissergeevitch/repo-task-proof-loop/claude-md"><img src="https://agentmods.dev/badge/instructions/denissergeevitch/repo-task-proof-loop/claude-md/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/denissergeevitch/repo-task-proof-loop/claude-md"><img src="https://agentmods.dev/badge/instructions/denissergeevitch/repo-task-proof-loop/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 424 This file is loaded in full into every session.

When invoked 424 The same file — it is already loaded in full.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00424	$0.00424
Opus 5.5	$0.00170	$0.00170
Sonnet 5.5	$0.00085	$0.00085
Haiku 4.5	$0.00042	$0.00042

Measured 2mo ago against content hash 5ef474c55512, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

repo-task-proof-loop CLAUDE.md 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 2mo 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.

CLAUDE.md · 34 lines

What it actually says

Repo task proof loop

For substantial features, refactors, and bug fixes, use the repo-task-proof-loop workflow.

Required artifact path:

Keep all task artifacts in .agent/tasks/<TASK_ID>/ inside this repository.

Required sequence:

Freeze .agent/tasks/<TASK_ID>/spec.md before implementation.
Implement against explicit acceptance criteria (AC1, AC2, ...).
Create evidence.md, evidence.json, and raw artifacts.
Run a fresh verification pass against the current codebase and rerun checks.
If verification is not PASS, write problems.md, apply the smallest safe fix, and reverify.

Hard rules:

Do not claim completion unless every acceptance criterion is PASS.
Verifiers judge current code and current command results, not prior chat claims.
Fixers should make the smallest defensible diff.

Installed workflow agents:

.claude/agents/task-spec-freezer.md
.claude/agents/task-builder.md
.claude/agents/task-verifier.md
.claude/agents/task-fixer.md

Claude Code note:

If init just created or refreshed these files during the current Claude Code session, do not assume the refreshed workflow agents are already available.
The main Claude session may auto-delegate to these workflow agents when the current proof-loop phase matches their descriptions. If automatic delegation is not precise enough, make the current proof-loop phase more explicit in natural language.
TodoWrite or the visible task/todo UI is optional session-scoped progress display only. The canonical durable proof-loop state is the repo-local artifact set under .agent/tasks/<TASK_ID>/.
Keep this workflow flat. These generated workflow agents are role endpoints, not recursive orchestrators.
Keep this block in the root CLAUDE.md. If the workflow needs longer repo guidance, prefer @path imports or .claude/rules/*.md instead of expanding this block.

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.

2mo ago First seen · 34 lines · 424 tokens per session scan A 5ef474c55512

Subscribe to this mod's changes

repo-task-proof-loop CLAUDE.md is an instructions file published in the GitHub repository DenisSergeevitch/repo-task-proof-loop (726 stars, last pushed 7mo ago), licensed Apache-2.0. It adds 424 tokens to every session, about $0.0017 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 2026-08-30.

repo-task-proof-loop CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
repo-task-proof-loop CLAUDE.mdDenisSergeevitch/repo-task-proof-loop	Instructions file	726	7mo ago	Apache-2.0	A	424	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

repo-task-proof-loop: Instructions file for Claude Code