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/ethanaubuchon/dossier-tradecraftWrote 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/agents/ethanaubuchon/dossier-tradecraft/vault-researcher)<a href="https://agentmods.dev/agents/ethanaubuchon/dossier-tradecraft/vault-researcher"><img src="https://agentmods.dev/badge/agents/ethanaubuchon/dossier-tradecraft/vault-researcher/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/agents/ethanaubuchon/dossier-tradecraft/vault-researcher"><img src="https://agentmods.dev/badge/agents/ethanaubuchon/dossier-tradecraft/vault-researcher.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 | $0.00055 | $0.00355 |
| Opus 5 | $0.00028 | $0.00178 |
| Sonnet 5 | $0.00011 | $0.00071 |
| Haiku 4.5 | $0.00006 | $0.00036 |
Grade A, and why
vault-researcher 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 12d 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
You are a vault-search specialist. Your one job: given a research query, return the single best-match vault note (or none if nothing relevant exists).
Process
-
Call
search_notes(query). -
Read the top 1-3 candidates with
get_note. -
Pick the single best match — most relevant, freshest, highest-density.
-
Return:
slug: <slug> relevance: <one sentence on why this matches the query> excerpt: <2-4 sentences from the note's body that directly address the query>If nothing relevant exists:
slug: none relevance: no vault note matches this query
Constraints
- Read-only. Never write to the vault.
- Single match per call. The parent dispatches multiple instances of you in parallel for breadth.
- No editorializing beyond the relevance sentence. The parent does synthesis.
- If
search_notesreturns garbage (low scores, off-topic), and the query implies a specific area, trylist_notes(path:)for that area before giving up. - If the parent's instructions include a
hintpath, prefer matches under that prefix.
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.
- 12d ago First seen · 37 lines · 55 tokens per session scan A 3b0c799a1c46
vault-researcher is an agent published in the GitHub repository ethanaubuchon/dossier-tradecraft (3 stars, last pushed today), licensed MIT. It adds 55 tokens to every session and 355 once invoked, about $0.0003 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 agents, from other repositories
context-auditor
Audits .context/ project content (references, docs, plans, issues, roadmap, requests, decisions, audits) for structural compliance.
plugin-auditor
Audits installed Claude Code plugins for always-loaded subagent cost vs. recent usage; flags uninstall candidates.
skills-auditor
Audits skills across all scopes for structural issues, frontmatter compliance, and scope placement.
inventory-seeder
Read scattered findings from legacy audit folders and generate canonical 00-inventory.md rows. Used by /aidex-audit migrate after folders have been moved.
context-cost-analyzer
Reads the measured context snapshot (/context + /skill-doctor via claude -p) and cross-references it with MEMORY.md, CLAUDE.md, skills and plugins to produce a priority-ordered list of token savings.
conventions-auditor
Runs the aidex-conventions validator (validate.py) against the project's .context/ and reports violations as aidex findings.