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.
npx agentmods add commands/odinmb/ops-workflow/tidygit clone --depth 1 https://github.com/OdinMB/ops-workflowWhat 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 | $0.00033 | $0.01873 |
| Opus 5 | $0.00016 | $0.00937 |
| Sonnet 5 | $0.00007 | $0.00375 |
| Haiku 4.5 | $0.00003 | $0.00187 |
Grade A, and why
tidy 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 2d 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 — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ops Housekeeping
Scan this project's knowledge structure for inconsistencies and staleness, fix everything that can be fixed automatically, then brief the user on the little that's left for them to decide.
Pre-check: Process .update/ Inbox
Before running consistency checks, process any pending files in .update/. Spawn the updater agent (agents/updater.md) to handle this in an isolated context. Pass it a follow-up file path at .tmp/housekeeping-update-followup.md.
If .update/ is empty or doesn't exist, skip this step.
Wait for the updater agent to finish before proceeding — the consistency checks should run against the freshly-updated knowledge base.
Phase 1: Scan
1. INDEX.md Accuracy
For every folder that contains an INDEX.md:
- Read the
INDEX.md. - List the actual files and subfolders in that directory.
- Flag:
- Missing entries: files/subfolders that exist but aren't listed in
INDEX.md - Stale entries: entries in
INDEX.mdthat reference files/subfolders that no longer exist - Outdated summaries: entries whose one-line description doesn't match the file's actual content (spot-check the 3 largest files per folder)
- Missing entries: files/subfolders that exist but aren't listed in
2. MEMORY.md References
- Read
MEMORY.md. - Check every file path or folder reference it mentions — verify they exist.
- Check that
state/file summaries in MEMORY.md roughly match the actual file content. - Flag references to deleted or renamed files.
3. Orphaned Files
Find .md files that aren't listed in any INDEX.md. Exclude:
MEMORY.md,CLAUDE.md,README.md(root-level files)- Files in
.claude/,.context/,.plans/directories - Files in
plans/(active plans aren't indexed)
4. Stale Dates
Check _Last updated:_ lines in state/ and backlog/ files. Flag any that are older than 60 days — they may need review.
5. WAITING Markers
Grep for WAITING ( across all files. For each match:
- Extract who/what is being waited on and the since-date
- Flag any that are older than 30 days — they may need follow-up
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.
- 2d ago First seen · 152 lines · 33 tokens per session scan A 49edf3f70897
tidy is a command published in the GitHub repository OdinMB/ops-workflow (7 stars, last pushed 15d ago), licensed MIT. It adds 33 tokens to every session and 1,873 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
kb-ingest
Esegui un ingest nella LLM Wiki per il progetto o argomento specificato: $ARGUMENTS.
kb-output
Esporta una pagina della LLM Wiki in formato slide deck Marp: $ARGUMENTS.
kb-promote
Promuovi il concetto specificato da llm-wiki/wiki/concepts/ alla KB stabile (questo repo o un fork): $ARGUMENTS.
kb-lint
Esegui un controllo qualità completo della LLM Wiki in $LLMWIKIPATH.
close
Close out a Linear issue — move to Done, add a closing comment, and clean up the local branch/worktree.
prompt
Convert a rough ask into a structured XML-tagged prompt using the factory-prompting vocabulary.