What is proofpress AGENTS.md?

A set of project instructions for Proofpress, a tool that tracks revisions to Markdown and static HTML knowledge files in a ledger.

Use it when editing tracked Markdown or static HTML knowledge artifacts, including capturing prior versions, preserving block identifiers, checking policy, and writing claims records.

It helps preserve human edits, keep files portable, and record exactly which documented claims changed. It prevents agents from treating every conversation as a versioned revision.

proofpress AGENTS.md — Instructions file by chenmingtang830

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 instructions/chenmingtang830/proofpress/agents-md

Clone the repo

git clone --depth 1 https://github.com/chenmingtang830/proofpress

Made for: Codex, OpenCode.

Per session 3,718 This file is loaded in full into every session.

When invoked 3,718 The same file — it is already loaded in full.

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.

Model	Per session	Once invoked
Fable 5	$0.03718	$0.03718
Opus 5	$0.01859	$0.01859
Sonnet 5	$0.00744	$0.00744
Haiku 4.5	$0.00372	$0.00372

Measured 2d ago against content hash 9331e14b05aa, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

proofpress AGENTS.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 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.

AGENTS.md · 43 lines

How it starts

The opening of the file, as written. The whole thing — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.

[//]: # (ob:e43408a0)
# Project instructions

[//]: # (ob:bbe579fd)
## Proofpress: ledger Markdown and static HTML knowledge artifacts

[//]: # (ob:34d3dbf3)
For accepted, meaningful revisions to Markdown or static HTML knowledge
artifacts—never source code—close this loop. Do not snapshot every
conversational turn.

[//]: # (ob:6c2c2b16)
Before editing an existing target, run
`python3 proofpress.py capture --recorder codex-preflight <file>`. This
preserves any human drift as a separate version without guessing its author or
reason. Then:

[//]: # (ob:1be95c90)
1. Read `python3 proofpress.py policy <file>`. Every tracked Markdown or static
   HTML knowledge artifact in this official repository must remain portable;
   run `policy <file> portable` if needed. The setting is sticky. Use `ingest
   <file>` for Git history not yet represented in the local ledger.
2. Preserve carrier-native block anchors while editing: Markdown uses invisible
   `ob` markers; static HTML uses `data-proofpress-id`. Then run `anchor <file>`
   and inspect inherited/new/gone IDs.
3. Write claims JSON with one honest item per touched or removed block. Do not
   enumerate untouched blocks.
4. Snapshot with `--why`, claims, and explicit actors (`requested_by`,
   `produced_by`, `edited_by`, `recorded_by`, and `attribution_basis` when
   known). Omit `--rejected` unless a consequential dead branch should stop
   future collaborators from repeating it. Never capture raw prompts or
   transcripts.
5. Run `verify <file>` and report its output verbatim. Never re-snapshot just to
   force green.

[//]: # (ob:32ab4de4)
For incoming portable files, run `inspect` and then `import`. Fallback `capture`
records only `recorded_by`; it cannot know authorship or reasoning.

[//]: # (proofpress:meta:eyJhcnRpZmFjdF9pZCI6InBwXzc5NWU0MTc0YjU2ZDdkNWNmODkxOTNlYSIsInBvbGljeSI6InBvcnRhYmxlIiwicG9ydGFibGVfaGVhZCI6IjAxYTM4YjczIiwicG9ydGFibGVfaGVhZF9ldmVudCI6InBwZV9mNWM4Njg3ZTJmYjE5MGYxNzQ4YTQ5MTQiLCJwb3J0YWJsZV9saW5lYW

Read the full file on GitHub · 43 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.

2d ago First seen · 43 lines · 3,718 tokens per session scan A 9331e14b05aa

Subscribe to this mod's changes

proofpress AGENTS.md is an instructions file published in the GitHub repository chenmingtang830/proofpress (5 stars, last pushed 2d ago), licensed Apache-2.0. It adds 3,718 tokens to every session, about $0.0186 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.

Writing and docs ai-agents audit-log documentation knowledge-management markdown provenance

proofpress AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

What this file has done since we first saw it

Other instructions, from other repositories

memory-forest AGENTS.md

core copilot-instructions.md

kipi-system AGENTS.md

scraps CLAUDE.md

data-olympus AGENTS.md

scratchpad CLAUDE.md