tri-party-framework: Command for Claude Code

.claude/commands/agentparty-claw.md

agentparty-claw is a command for Claude Code from r-design-j/tri-party-framework. It costs 17 tokens per session (879 once invoked), scanned A, original, MIT.

A command for creating or inspecting a handoff kit between Claude Code and Feishu Claw, an agent collaboration environment. It uses an existing AgentParty framework to coordinate that handoff.

In plain words
What is it for?
It is for checking AgentParty status and dispatching tasks through the installed framework, including its setup checks and handoff inspection.
Why use it?
It removes the need to manually assemble or bypass the project's handoff protocol.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions Claude Code; mentions Codex.

This is r-design-j/tri-party-framework's own configuration. It tells Claude Code how to work on tri-party-framework 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 tri-party-framework configures →

Reuse

Borrowing it

Nothing to install: this file belongs to r-design-j/tri-party-framework. 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/r-design-j/tri-party-framework/main/.claude/commands/agentparty-claw.md
Clone the repo
git clone --depth 1 https://github.com/r-design-j/tri-party-framework

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 agentparty-claw

README.md
[![agentmods](https://agentmods.dev/badge/commands/r-design-j/tri-party-framework/agentparty-claw/github.svg)](https://agentmods.dev/commands/r-design-j/tri-party-framework/agentparty-claw)
Your own site
<a href="https://agentmods.dev/commands/r-design-j/tri-party-framework/agentparty-claw"><img src="https://agentmods.dev/badge/commands/r-design-j/tri-party-framework/agentparty-claw/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 agentparty-claw

Your own site · 80×15
<a href="https://agentmods.dev/commands/r-design-j/tri-party-framework/agentparty-claw"><img src="https://agentmods.dev/badge/commands/r-design-j/tri-party-framework/agentparty-claw.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 879 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 $0.00017 $0.00879
Opus 5 $0.00009 $0.00439
Sonnet 5 $0.00003 $0.00176
Haiku 4.5 $0.00002 $0.00088

Measured 8d ago against content hash 752ae699c81d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

agentparty-claw 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 8d 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/commands/agentparty-claw.md · 64 lines

What it actually says

Use the existing AgentParty portable framework. Do not recreate the protocol by writing new ad hoc Markdown files.

Arguments:

$ARGUMENTS

Resolve the command in this order:

  1. agentparty from PATH.
  2. $AGENTPARTY_FRAMEWORK_HOME/scripts/agentparty.sh after reading ~/.triparty-framework/config.
  3. scripts/agentparty.sh from the current repository.

Run this resolver before dispatch:

AGENTPARTY_CMD="$(command -v agentparty 2>/dev/null || true)"
if [ -z "$AGENTPARTY_CMD" ] && [ -f "$HOME/.triparty-framework/config" ]; then
  . "$HOME/.triparty-framework/config"
  if [ -n "${AGENTPARTY_FRAMEWORK_HOME:-}" ] && [ -x "$AGENTPARTY_FRAMEWORK_HOME/scripts/agentparty.sh" ]; then
    AGENTPARTY_CMD="$AGENTPARTY_FRAMEWORK_HOME/scripts/agentparty.sh"
  fi
fi
if [ -z "$AGENTPARTY_CMD" ] && [ -x "scripts/agentparty.sh" ]; then
  AGENTPARTY_CMD="scripts/agentparty.sh"
fi
if [ -z "$AGENTPARTY_CMD" ]; then
  printf '%s\n' "AgentParty is not installed or not discoverable in this environment." >&2
  printf '%s\n' "After installation, run: agentparty doctor" >&2
  printf '%s\n' "Install or clone: https://github.com/r-design-j/tri-party-framework.git" >&2
  exit 1
fi

If all three resolution paths fail, stop with the resolver's non-zero exit. Report that AgentParty is not installed or not discoverable in the current environment, suggest running agentparty doctor after installation, and ask whether to clone or install https://github.com/r-design-j/tri-party-framework.git. Do not create replacement protocol files, placeholder Markdown, or local ad hoc workflows.

Dispatch:

  • Empty arguments: run "$AGENTPARTY_CMD" guide --pack claude-code-feishu-claw.
  • prompt <task>: run "$AGENTPARTY_CMD" prompt --pack claude-code-feishu-claw --task "<task>".
  • kit <task>: run "$AGENTPARTY_CMD" kit --pack claude-code-feishu-claw --task "<task>" --out "agentparty-claw-kit-$(date +%Y%m%d-%H%M%S)".
  • guide <kit-dir-or-run-dir>: run "$AGENTPARTY_CMD" guide --pack claude-code-feishu-claw --run-dir "<kit-dir-or-run-dir>".
  • validate <kit-dir-or-run-dir>: run "$AGENTPARTY_CMD" validate-run --run-dir "<kit-dir-or-run-dir>".
  • Otherwise treat $ARGUMENTS as the task and create a new local handoff kit with "$AGENTPARTY_CMD" kit --pack claude-code-feishu-claw --task "$ARGUMENTS" --out "agentparty-claw-kit-$(date +%Y%m%d-%H%M%S)".

Report the kit directory, state.json, evidence bundle path, next command, and boundaries.

Boundaries:

  • This is a 2-agent AgentParty product pack, not Codex + Claude + Gemini true tri-party.
  • This pack does not participate in Codex + Claude + Gemini cross-audit unless a separate triparty run is explicitly started.
  • Do not set or claim true_triparty_ready=true.
  • The slash layer must not call Feishu, configure Claw auth, import evidence, or claim pack_ready by itself.
  • Evidence import remains a WSL2/macOS/Linux executable path; native PowerShell is preparation-only.
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. 8d ago First seen · 64 lines · 17 tokens per session scan A 752ae699c81d

Subscribe to this mod's changes

agentparty-claw is a command published in the GitHub repository r-design-j/tri-party-framework (4 stars, last pushed 2mo ago), licensed MIT. It adds 17 tokens to every session and 879 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.