What is task-agent CLAUDE.md?

A Claude Code instruction file that imports the project’s shared rules from their canonical files. Claude Code is a programming assistant that follows repository instructions while working on code.

Use it to connect Claude Code with the project’s rules for bootstrapping, context discovery, Git synchronization and safety, and task updates.

It avoids copying the same rules into multiple agent-specific files, which could otherwise become inconsistent.

task-agent CLAUDE.md — Instructions file by rdudov

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/rdudov/task-agent/claude-md

Clone the repo

git clone --depth 1 https://github.com/rdudov/task-agent

Per session 698 This file is loaded in full into every session.

When invoked 698 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	$0.00698	$0.00698
Opus 5	$0.00349	$0.00349
Sonnet 5	$0.00140	$0.00140
Haiku 4.5	$0.00070	$0.00070

Measured 2d ago against content hash 42cf1aa1bdb0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

task-agent 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 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.

CLAUDE.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Claude Code Entry Point

This file exists so Claude Code loads the same operating rules that Codex and Cursor already use. It deliberately contains no rules of its own: every rule below is imported from its canonical file, so there is exactly one source of truth per rule and nothing is duplicated per agent.

Project rules

@AGENTS.md

Always-on rules

These are the Cursor rules. Each path below is a symlink to the canonical .cursor/rules/*.mdc file; the content lives there and only there.

@.claude/imports/task-bootstrap-required.md @.claude/imports/context-discovery.md @.claude/imports/git-remote-sync.md @.claude/imports/git-push-safety.md @.claude/imports/task-artifact-updates.md

Repository wiring for Claude Code

AGENTS.md is the canonical project rule file, shared with Codex. It is imported above rather than copied.
.cursor/rules/*.mdc are the canonical rule files, shared with Cursor. They are imported above through .claude/imports/*.md symlinks rather than copied. Claude Code applies all of them on every session; Cursor additionally scopes task-artifact-updates by its globs frontmatter, which Claude Code ignores.
skills/ is the canonical skill library. It is exposed to Claude Code through the .claude/skills symlink, so repository skills appear as normal Agent Skills. Invoke them by name instead of copying their instructions.
Environment-specific rules do not belong in this committed file. Keep them in a gitignored CLAUDE.local.md.

Edit the canonical files, never the symlinks, and never move a canonical file into .claude/. Codex reads AGENTS.md from the repository root, and Cursor reads .cursor/rules/.

.claude/imports/ exists only because Claude Code imports files by the .md extension and silently ignores an @….mdc import. Import resolution does follow symlinks, so a .md symlink to the .mdc original carries the rule without duplicating it. The directory holds nothing but those symlinks.

docs/claude-code-setup.md explains how to recreate, change, or verify this wiring.

Read the full file on GitHub · 62 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 · 62 lines · 698 tokens per session scan A 42cf1aa1bdb0

Subscribe to this mod's changes

task-agent CLAUDE.md is an instructions file published in the GitHub repository rdudov/task-agent (4 stars, last pushed 2d ago), licensed MIT. It adds 698 tokens to every session, about $0.0035 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.

Building agents Writing and docs

task-agent CLAUDE.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Claude Code Entry Point

Project rules

Always-on rules

Repository wiring for Claude Code

What this file has done since we first saw it

Other instructions, from other repositories

spellbook AGENTS.md

open-supermarkets AGENTS.md

helm copilot-instructions.md

flyto-core CLAUDE.md

fusion-skills AGENTS.md

aeon CLAUDE.md