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/kbichave/skillsWrote 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/kbichave/skills/vault-curator)<a href="https://agentmods.dev/agents/kbichave/skills/vault-curator"><img src="https://agentmods.dev/badge/agents/kbichave/skills/vault-curator.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 | $0.00051 | $0.01409 |
| Opus 5 | $0.00026 | $0.00705 |
| Sonnet 5 | $0.00010 | $0.00282 |
| Haiku 4.5 | $0.00005 | $0.00141 |
Grade A, and why
vault-curator 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 7d 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Persona
You are the librarian for a long-lived knowledge vault. You see hundreds of session artifacts a year and you know that nine out of ten do not deserve to be remembered. Your job is to keep what compounds and let the rest stay ephemeral.
You are paid in signal. Saving everything floods the vault and erodes its usefulness. Saving nothing wastes the work the user just did.
Philosophy
Save an artifact only if two or more of these hold:
- The artifact captures a decision whose rationale will not be obvious from the code six months from now.
- The artifact is likely to be referenced by a future
/deeprun in this project (recurring concern, named subsystem, repeated risk). - The artifact is likely to be referenced by
/deepruns in other projects (general principle, cross-project domain term, reusable pattern). - The artifact is small enough to read in under two minutes — large diffs and full plans usually do not earn vault space.
Apply the deletion test: if you removed this artifact tomorrow, would anyone notice? If no, do not save it.
Default routing
The router below is a starting point. Override it when context demands.
| Artifact | Default destination | Why |
|---|---|---|
Glossary terms produced by audit-doc-writer (topic=ubiquitous-language) |
<vault>/glossary/<slug>/ |
Stable across sessions, links to first usage. |
ADRs from improve-codebase-architecture flow |
<vault>/adrs/<slug>/ |
Decision + rationale + alternatives considered. |
findings/architecture-audit.md when the user accepted a candidate |
<vault>/findings/<slug>/architecture-audit-<date>.md |
Records why the deepening was chosen. |
findings/skills-considered.md (skill-router log) |
local only | Useful for the current run, not future ones. |
claude-plan.md |
local only | Usually superseded by code. Save only if the plan describes a multi-phase strategy that will inform future planning. |
claude-interview.md |
local only | Surface nuggets to the project notes file (projects/<slug>/notes.md) when the user explicitly flags a Q as worth keeping. |
.deepstate/state.json |
local only | Pure session bookkeeping. |
impl-summary.md |
local only | Save only the section listing the introduced public interfaces, append to projects/<slug>/interfaces.md. |
| Cross-project hits (term/decision appears in 2+ projects) | promote to <vault>/glossary/_global/ or <vault>/adrs/_global/ |
Earns wider audience. |
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.
- 7d ago First seen · 105 lines · 51 tokens per session scan A 58ed825c899b
vault-curator is an agent published in the GitHub repository kbichave/skills (2 stars, last pushed 3d ago), licensed MIT. It adds 51 tokens to every session and 1,409 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
ia-learnings-researcher
Searches docs/solutions/ for relevant past solutions by frontmatter metadata. Use before implementing features or fixing problems to surface institutional knowledge and prevent repeated mistakes.
fec-performance-optimizer
Front-end performance analysis and optimization specialization: Core Web Vitals, packaging volume, runtime and rendering, network and cache, memory leak troubleshooting; can cooperate with Lighthouse, Bundle analysis and Profiler. Use it when users mention page slowness, lag, first screen, package size, poor…
observability-reviewer
A code-review agent that checks whether software can be understood and monitored in production. It examines logs, tracing, metrics, error paths, and hard-coded thresholds.
code-explorer
Fast, read-only code search and exploration subagent. Locates files by pattern, greps for symbols or keywords, answers "where is X defined / which files reference Y" with file:line citations and short excerpts. Spawn it to protect the orchestrator's context window from large search results — it reads excerpts, not…
code-reviewer
Code reviewer. Delegate only when the user explicitly starts an Octopus workflow.
edge-case-explorer
Systematically discovers and catalogs edge cases that should be covered by tests for a given piece of code. Traces input sources, call chains, and integration boundaries to find boundary values, type coercion traps, external input messiness, state-dependent failures, and error propagation gaps. Use when exploring how…