audit

An append-only record of agent prompts, replies, tool calls, errors, and review labels in a JSONL file. JSONL is a text format with one JSON record per line.

In plain words
What is it for?
Use it to record model and tool interactions, associate failures with issues, and label past results as good or bad with reasons.
Why use it?
It preserves a history of agent activity without changing earlier records.

Command

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 commands/gastownhall/beads/audit
Clone the repo
git clone --depth 1 https://github.com/gastownhall/beads
Per session 11 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 256 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 $0.00011 $0.00256
Opus 5 $0.00005 $0.00128
Sonnet 5 $0.00002 $0.00051
Haiku 4.5 $0.00001 $0.00026

Measured yesterday against content hash bf941ae72f3d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

audit 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 yesterday.

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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

  • audit — 89% identical, 5 lines differ
plugins/beads/skills/beads/commands/audit.md · 28 lines

What it actually says

Append-only audit logging for agent interactions (prompts, responses, tool calls) in .beads/interactions.jsonl.

Each line is one event. Labeling is done by appending a new "label" event referencing a previous entry.

Usage

  • Record an interaction:

    • bd audit record --kind llm_call --model "claude-3-5-haiku" --prompt "..." --response "..."
    • bd audit record --kind tool_call --tool-name "go test" --exit-code 1 --error "..." --issue-id bd-42
  • Pipe JSON via stdin:

    • cat event.json | bd audit record
  • Label an entry:

    • bd audit label int-a1b2 --label good --reason "Worked perfectly"
    • bd audit label int-a1b2 --label bad --reason "Hallucinated a file path"

Notes

  • Audit entries are append-only (no in-place edits).
  • bd dolt push includes .beads/interactions.jsonl in the commit allowlist.
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. yesterday First seen · 28 lines · 11 tokens per session scan A bf941ae72f3d

Subscribe to this mod's changes

audit is a command published in the GitHub repository gastownhall/beads (26,754 stars, last pushed yesterday), licensed MIT. It adds 11 tokens to every session and 256 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.