forensics

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/tatargabor/set-copilot/forensics

Clone the repo

git clone --depth 1 https://github.com/tatargabor/set-copilot

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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/tatargabor/set-copilot/forensics.svg)](https://agentmods.dev/commands/tatargabor/set-copilot/forensics)

Your own site

<a href="https://agentmods.dev/commands/tatargabor/set-copilot/forensics"><img src="https://agentmods.dev/badge/commands/tatargabor/set-copilot/forensics.svg" alt="Measured on agentmods" height="20"></a>

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 819 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.

Model	Per session	Once invoked
Fable 5	$0.00000	$0.00819
Opus 5	$0.00000	$0.00409
Sonnet 5	$0.00000	$0.00164
Haiku 4.5	$0.00000	$0.00082

Measured today against content hash 32faa7ce0bd1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

forensics 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 today.

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/set/forensics.md · 72 lines

How it starts

The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Forensic analysis of a completed orchestration run.

Input: <run-id> (the e2e-runs directory name, e.g. craftbrew-run-20260421-0025). If omitted, ask the user which run to analyse.

You are debugging a completed orchestration run. Use set-run-logs to surface filtered views over Claude Code session transcripts + orchestration logs without dragging raw 300+ KB jsonl files into context.

CRITICAL: never read raw session jsonl unbounded

A single session jsonl can exceed 300 KB; a run has 50+ sessions. Always use the CLI's filtered views FIRST. Only fall back to Read on a specific line range after the CLI has narrowed the suspect.

Triage flow (follow in order)

Discover:
```
set-run-logs <run-id> discover
```
Confirms which session dirs (main + each worktree) and orchestration artifacts were found.
Digest — get the error map:
```
set-run-logs <run-id> digest
```
Aggregates tool_error, bash_nonzero_exit, stop_reason_anomaly, user_interrupt, permission_denial, and crash_suspect signals across every session, grouped by change → session → tool with snippets and counts.
Drill into suspect sessions — for each suspect from the digest:
```
set-run-logs <run-id> session <uuid-prefix> --errors-only
```
The UUID prefix from the digest (e.g. 531d481b) is enough — minimum 6 chars. Use --tool <Bash|Edit|Read|...> to narrow further.
Targeted probes — when a specific error signature matters:
```
set-run-logs <run-id> grep "<pattern>"
set-run-logs <run-id> grep "EACCES" --tool Bash
set-run-logs <run-id> grep "MaxToken" -i --limit 100
```
Emits only the matching message.content text, never raw jsonl. Default cap is 50 matches — raise with --limit only when you must.
Cross-reference orchestration timing:
```
set-run-logs <run-id> orchestration
```
Per-change dispatch counts, last gate status, terminal status (e.g. MERGE_SUCCESS, FIX_ISS_ESCALATED), gate-outcome totals. Use this to map a suspect session back to the engine's view.

Read the full file on GitHub · 72 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.

today First seen · 72 lines · 0 tokens per session scan A 32faa7ce0bd1

Subscribe to this mod's changes

forensics is a command published in the GitHub repository tatargabor/set-copilot (2 stars, last pushed 12d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 819 tokens. 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 2026-09-04.

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

CRITICAL: never read raw session jsonl unbounded

Triage flow (follow in order)

What this file has done since we first saw it

Other commands, from other repositories

checklist

clarify

specify

analyze

converge

implement