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/Marazii/research-co-pilotWrote 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/marazii/research-co-pilot/vault)<a href="https://agentmods.dev/commands/marazii/research-co-pilot/vault"><img src="https://agentmods.dev/badge/commands/marazii/research-co-pilot/vault.svg" alt="Measured on agentmods" 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 | $0.00021 | $0.00387 |
| Opus 5 | $0.00010 | $0.00193 |
| Sonnet 5 | $0.00004 | $0.00077 |
| Haiku 4.5 | $0.00002 | $0.00039 |
Grade A, and why
vault 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 8d 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.
What it actually says
Invoke the vault skill from the research-co-pilot plugin and execute its full workflow.
The skill file is at skills/vault/SKILL.md relative to this plugin. Read it and follow it precisely — including:
- The PII hard rule (the vault holds pseudonyms only; never real names / emails / phones / addresses).
- Files are the source of truth; the vault is reconciled, never trusted blindly.
- Distinct facts (recruited-N vs analyzed-N) are kept as separate keys, not overwritten.
- Subcommands:
show(default) /init/audit/add <type>/resolve <question>. initscaffolds the visibleresearch/<slug>/folder template (numbered stage folders +knowledge/+audits/+archive/+ manifest + root README).organizefiles loose artifacts into their stage folders.auditruns the seven checks (fact consistency, citation consistency, terminology, open questions, staleness, PII safety, filing) and writesaudits/vault_audit_<date>.mdwith 🔴/🟡/🟢 findings.organizemoves files andauditreports — neither edits an artifact's content.
The vault spec is docs/research-vault.md.
User input: $ARGUMENTS
If no subcommand was given, default to show. If there's no project vault yet (no research/<slug>/manifest.json), say so and offer to run init.
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.
- 8d ago First seen · 21 lines · 21 tokens per session scan A 703cc7f30f06
vault is a command published in the GitHub repository Marazii/research-co-pilot (13 stars, last pushed 3mo ago), licensed MIT. It adds 21 tokens to every session and 387 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.
Other commands, from other repositories
setup-context
Set up your Skill Memory — a pm-context.md every skill reads so outputs come back tailored to you.
brain
Set up or use your Professional Brain — init, ingest, recall, or review.
session-learn
Extract knowledge from the current session. Invoke with /session-learn or "extract this session".
memory-forget
Command "memory-forget" from waittim/MemoryCustodian, covering memory-forget and review the preview, then apply its plan id.
reflect
Capture learnings from the current conversation through guided reflection.
memory-compact
Run a dry run first.