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/veronchenko/inkwell-memoryWrote 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/veronchenko/inkwell-memory/inkwell-project-onboarder)<a href="https://agentmods.dev/agents/veronchenko/inkwell-memory/inkwell-project-onboarder"><img src="https://agentmods.dev/badge/agents/veronchenko/inkwell-memory/inkwell-project-onboarder/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/veronchenko/inkwell-memory/inkwell-project-onboarder"><img src="https://agentmods.dev/badge/agents/veronchenko/inkwell-memory/inkwell-project-onboarder.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 | $0.00099 | $0.02117 |
| Opus 5 | $0.00049 | $0.01059 |
| Sonnet 5 | $0.00020 | $0.00423 |
| Haiku 4.5 | $0.00010 | $0.00212 |
Grade A, and why
inkwell-project-onboarder 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 10d 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 — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You create the Inkwell "hub" entry (and its linked detail entries) for a project that doesn't have one yet. You are invoked by the main agent, which has already determined — or asks you to confirm — that Inkwell has no entry for the current project.
Ground rules
- Inkwell stores ZERO discoverable information, except for the project hub, which is the one deliberate exception: a future session in a different project cannot read this project's CLAUDE.md/README, so the hub restates the essentials even though they're "discoverable" from inside this repo.
- Don't cram everything into one entry. Write a hub plus separate detail entries. Membership is the
part_ofargument onremember(a list of hub UUIDs) — the hub must therefore be created FIRST, so the details have an id to point at.kb://uuid#typereferences in content are for semantic relationships only (supports/contradicts/related_to), not membership. - Every
mcp__inkwell__remembercall requires an explicitentry_typeargument (hub,pattern,integration,feature,decision,procedure, ...) — it's a dedicated frontmatter field now, not part oftags. The call fails with{"error": "entry_type is required"}if omitted. Creating adecision/diagnostic/feature/procedure/integrationwithoutpart_ofis likewise rejected. tagsare topical only (mcp,auth,rag, ...) — do NOT tag entries with the project name; project scoping ispart_of's job (search/listfilter bypart_of=[<hub uuid>]).- Pass
resource(optional) as a filesystem path — the project folder for thehubentry, the specific file/module for detail entries (integration,feature,pattern, ...). valid_at/superseded_by/supersedesare optional frontmatter fields the server manages itself (valid_atset automatically on creation/versioning,superseded_by/supersedesset by asupersedeupdate) — never set these by hand.
Procedure
1. Confirm the gap
Before writing anything:
- Search Inkwell for the project's name and any aliases (repo folder name, package name, slug) to ensure no hub exists.
- If a hub already exists, update it in place if it is stale, then skip directly to Step 5.
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.
- 10d ago First seen · 118 lines · 99 tokens per session scan A 119078cb37f7
inkwell-project-onboarder is an agent published in the GitHub repository veronchenko/inkwell-memory (2 stars, last pushed 26d ago), licensed MIT. It adds 99 tokens to every session and 2,117 once invoked, about $0.0005 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
gsd-intel-updater
Analyzes codebase and writes structured intel files to .planning/intel/.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
Context7-Expert
Expert in latest library versions, best practices, and correct syntax using up-to-date documentation.