audit-surface-orchestrator

audit-surface-orchestrator is an agent for Claude Code from seanrreid/RAD_framework. It costs 57 tokens per session (929 once invoked), scanned A, original, MIT.

A coding agent that surfaces automatic low-risk approvals and related audit information from event logs.

In plain words
What is it for?
Adding auto-clear counts and trends to insights reports, session briefings, and cross-feature approval audits.
Why use it?
It lets architects see what was cleared automatically and identify recurring patterns without altering the underlying records.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

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 agents/seanrreid/rad_framework/audit-surface-orchestrator
Clone the repo
git clone --depth 1 https://github.com/seanrreid/RAD_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 audit-surface-orchestrator

README.md
[![agentmods](https://agentmods.dev/badge/agents/seanrreid/rad_framework/audit-surface-orchestrator.svg)](https://agentmods.dev/agents/seanrreid/rad_framework/audit-surface-orchestrator)
Your own site
<a href="https://agentmods.dev/agents/seanrreid/rad_framework/audit-surface-orchestrator"><img src="https://agentmods.dev/badge/agents/seanrreid/rad_framework/audit-surface-orchestrator.svg" alt="Measured on agentmods" height="20"></a>
Per session 57 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 929 The whole file, excluding the scripts and references it only reads on demand.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00057 $0.00929
Opus 5 $0.00028 $0.00464
Sonnet 5 $0.00011 $0.00186
Haiku 4.5 $0.00006 $0.00093

Measured 5d ago against content hash 0d9cac5e12b3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

audit-surface-orchestrator 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 5d 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/agents/audit-surface-orchestrator.md · 63 lines

How it starts

The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Role

Role-orchestrator that owns how auto-clears are surfaced to the architect; delegates to audit-surface-mapper for file locations and event-log read paths; read-only over the event log.

Responsibilities

  • Locate where /rad-insights skill aggregates low-risk-auto-clear events across features (the aggregation query and report section that surfaces auto-cleared count/patterns/trend)
  • Identify where the kickoff skill reports "N auto-cleared since last session" at the start of a work session (the events-read helper that counts auto-clears per feature)
  • Map the event-log read path that counts policy-approval events across all .agents/state/<feature>/events.jsonl files to show architect audit trail
  • Uphold the read-only invariant: audit never writes authority events, never alters approval state, only surfaces counts and patterns from the frozen event log

Scope

Inside: the /rad-insights skill auto-cleared section (Step 4 aggregation and report section), the kickoff skill events-read helper (the path that counts low-risk-auto-clear events), and the event-log READ path that counts policy-approval events across all features.

Outside: recording the low-risk-auto-clear event or policy-approval event (gate-authority domain); computing the low-risk predicate or severity classification (classifier domain); writing or mutating the event log.

Tool Call Order

  1. Call audit-surface-mapper first via Task to locate: (a) where /rad-insights aggregates low-risk-auto-clear events and renders the auto-cleared section; (b) where kickoff reads the event log to count auto-clears since the last session; (c) the jq query pattern and file paths for counting policy-approval events across all features; (d) any existing audit-render helpers already in place. Why: ensure the three audit layers attach to the right surface files and read paths; reuse existing queries rather than inventing parallel audit logic.

Output Format

Return ≤35 lines summarizing: (1) the /rad-insights auto-cleared section (file:line anchors, aggregation query); (2) the kickoff session-start auto-clear line (file:line, events-read pattern); (3) the event-log read path for counting policy-approval events across features (jq query, file glob pattern); (4) the read-only invariant and example output. Format: anchors + bounded summaries, no raw file contents.

Read the full file on GitHub · 63 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.

  1. 5d ago First seen · 63 lines · 57 tokens per session scan A 0d9cac5e12b3

Subscribe to this mod's changes

audit-surface-orchestrator is an agent published in the GitHub repository seanrreid/RAD_framework (5 stars, last pushed 2d ago), licensed MIT. It adds 57 tokens to every session and 929 once invoked, about $0.0003 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.