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.
git clone --depth 1 https://github.com/Ertinox7711/SGRR-AGI-V2Wrote 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.
[](https://agentmods.dev/commands/ertinox7711/sgrr-agi-v2/rig-audit)<a href="https://agentmods.dev/commands/ertinox7711/sgrr-agi-v2/rig-audit"><img src="https://agentmods.dev/badge/commands/ertinox7711/sgrr-agi-v2/rig-audit/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.
<a href="https://agentmods.dev/commands/ertinox7711/sgrr-agi-v2/rig-audit"><img src="https://agentmods.dev/badge/commands/ertinox7711/sgrr-agi-v2/rig-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 5.1 | $0.00030 | $0.00843 |
| Opus 5 | $0.00015 | $0.00421 |
| Sonnet 5 | $0.00006 | $0.00169 |
| Haiku 4.5 | $0.00003 | $0.00084 |
Grade A, and why
rig-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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are running the SGRR AGI V2 self-audit. Goal: study how this user actually works, compare it to what the rig currently provides, and produce a prioritized proposal of upgrades.
This run is report-only. You MUST NOT modify settings, CLAUDE.md, memory, rules, commands, hooks, or any file. Propose; let the user choose what to apply.
Optional focus from the user: $ARGUMENTS (if empty, audit everything).
Work through these five steps, then output the report.
1. Read the current rig
Read what is actually installed under the user's ~/.claude (Windows: $env:USERPROFILE\.claude):
settings.json— model, env, permissions, hooks, enabledPlugins, effortLevelCLAUDE.md— behavioral philosophyrules/*.md— lazy per-project rules and theirpaths:memory/MEMORY.md+ memory filescommands/*.md— existing slash commandsscripts/*— existing hooks
Build a model of what the rig CAN do today.
2. Sample real usage (do NOT ingest everything)
Session transcripts live in ~/.claude/projects/<encoded-path>/*.jsonl (one JSON object per line, files can be huge). Sample, don't load whole files — list the dirs, pick a handful of recent/large transcripts, read only their first and last slices. Extract signal only:
- recurring intents / task types the user asks for
- tools and skills used most (and ones never used)
- friction: repeated corrections, retries, the same thing explained twice, dead ends
Never copy secrets, tokens, emails, or private project content into your output. Summarize patterns, not raw data.
3. Scan the project folders
For each working directory (see additionalDirectories in settings, plus the cwd), survey:
- languages / frameworks / tooling in use
- candidate
paths:rules that don't exist yet - commands the user runs by hand repeatedly (build, test, deploy) that could be a slash command or a hook
4. Cross-reference and find the gaps
Compare step 1 against steps 2+3. Look for:
- missing rules — a project worked on often with no
rules/entry - missing memory — facts re-explained across sessions that should be persisted
- missing commands — a repeated multi-step flow worth a
/command - useful skills/plugins not enabled
- settings drift — permissions too tight or too loose, model/effort mismatch, a hook that never fires
- stale references — a rule/memory naming a file/flag that no longer exists (verify before flagging)
- new hook opportunities — a check the user always does by hand
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.
- 11d ago First seen · 62 lines · 30 tokens per session scan A 4f2ce1818e40
rig-audit is a command published in the GitHub repository Ertinox7711/SGRR-AGI-V2 (1 stars, last pushed 3d ago), licensed MIT. It adds 30 tokens to every session and 843 once invoked, about $0.0002 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.
Other commands, from other repositories
evolve
Cluster related instincts into skills, commands, or agents.
suggest-automation
A command that reviews recent Git commits and project instructions to find work patterns that happen repeatedly.
instinct-import
Import instincts from teammates, Skill Creator, or other sources.
instinct-export
Export instincts for sharing with teammates or other projects.
handover
Generate comprehensive context transfer document for agent session handoffs.
rewind-summary
Capture a carry-forward summary of work done since a checkpoint so it survives a native rewind/Restore. Run before rewinding.