p2p-feedback

p2p-feedback is a command for Claude Code from sanic732/P2P-4PDA-edition. It costs 13 tokens per session (281 once invoked), scanned A, original, MIT.

A command for recording whether an agent’s result was good, acceptable, or bad, then updating routing history and session metrics.

In plain words
What is it for?
Use it after a task to record quality, optionally name the agent, and update its routing bias and correction count.
Why use it?
It turns outcome feedback into data that can influence future agent selection and expose corrections or quality changes.

Command for Claude Code

Written for Claude Code: argument-hint in frontmatter.

Part of the p2p-v8c3 plugin — 9 skills, 13 commands, 8 agents, 2 hooks shipped together

Good fit Use it after a task to record quality, optionally name the agent, and update its routing bias and correction count.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/sanic732/p2p-4pda-edition/p2p-feedback
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.

Clone the repo
git clone --depth 1 https://github.com/sanic732/P2P-4PDA-edition

Made for: Claude Code.

Or install p2p-v8c3, the plugin that ships this one along with the rest of its 9 skills, 13 commands, 8 agents, 2 hooks.

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 p2p-feedback

README.md
[![agentmods](https://agentmods.dev/badge/commands/sanic732/p2p-4pda-edition/p2p-feedback/github.svg)](https://agentmods.dev/commands/sanic732/p2p-4pda-edition/p2p-feedback)
Your own site
<a href="https://agentmods.dev/commands/sanic732/p2p-4pda-edition/p2p-feedback"><img src="https://agentmods.dev/badge/commands/sanic732/p2p-4pda-edition/p2p-feedback/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 p2p-feedback

Your own site · 80×15
<a href="https://agentmods.dev/commands/sanic732/p2p-4pda-edition/p2p-feedback"><img src="https://agentmods.dev/badge/commands/sanic732/p2p-4pda-edition/p2p-feedback.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 13 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 281 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 $0.00013 $0.00281
Opus 5 $0.00006 $0.00140
Sonnet 5 $0.00003 $0.00056
Haiku 4.5 $0.00001 $0.00028

Measured 10d ago against content hash 73659e285b32, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

p2p-feedback 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 10d 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.

editions/8.4.6-C/plugin/.claude/commands/p2p-feedback.md · 32 lines

What it actually says

/p2p-feedback — Обратная Связь

Что делает: Записывает оценку результата, обновляет routing memory и metrics.

Использование:

/p2p-feedback good     → quality_score=1.0, agent_bias+10%
/p2p-feedback ok       → quality_score=0.7
/p2p-feedback bad      → quality_score=0.3, agent_bias-15%, corrections++
/p2p-feedback [агент] good/bad → точечное обновление routing memory

Пример: /p2p-feedback TECTON good → TECTON bias +10%

======================================== FILE_META

id: CMD_FEEDBACK_V8C type: command edition: CLAUDE_NATIVE invariants_passed: [I1_yaml, I2_api_strings, I3_deadlines, I4_g_errors, I5_version_metadata, I6_xml_native, I7_agents_8]

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. 10d ago First seen · 32 lines · 13 tokens per session scan A 73659e285b32

Subscribe to this mod's changes

p2p-feedback is a command published in the GitHub repository sanic732/P2P-4PDA-edition (17 stars, last pushed 22d ago), licensed MIT. It adds 13 tokens to every session and 281 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 2026-08-30.