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 skills/dentiny/kon/memory-loadingnpx skills add dentiny/kon --skill memory-loadinggit clone --depth 1 https://github.com/dentiny/konWhat 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 | $0.00058 | $0.00999 |
| Opus 5 | $0.00029 | $0.00500 |
| Sonnet 5 | $0.00012 | $0.00200 |
| Haiku 4.5 | $0.00006 | $0.00100 |
Grade A, and why
memory-loading 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 yesterday.
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 — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Memory Loading
Owner: all memory-reading agents
Consumers: agents/Azusa.md, agents/Mugi.md, agents/Mio.md, agents/Jun.md
Core principles (always): follow skills/core-principles — load only entries relevant to the task; don't hide gaps in memory coverage.
Storage layout
| Scope | Path | Contents |
|---|---|---|
| Repo skills (always first) | ~/.kon/projects/<repo-name>/skills/*/ |
Your named skill dirs for this repo — loaded verbatim before memory entries |
| Public (cross-project) | ~/.kon/public/memory/ |
User prefs, habits, feedback across repos |
| Repo (per project) | ~/.kon/projects/<repo-name>/memory/ |
Conventions specific to this checkout |
Each scope has a MEMORY.md index plus one file per entry (<slug>.md).
Resolve paths when documenting commands:
python3 $KON_ROOT/hooks/_kon_paths.py public-memory
python3 $KON_ROOT/hooks/_kon_paths.py project-memory
Override data root with KON_DATA_DIR (default ~/.kon).
Standard load flow
Before starting work:
- Load repo skills — list files with
python3 $KON_ROOT/hooks/_kon_paths.py project-skill-files(returns all~/.kon/projects/<repo-name>/skills/*/SKILL.mdsorted by skill name). Read every listed file in full and treat as the first context for all downstream work. Print[repo-skill: loaded — build, conventions, ...]or[repo-skill: not found]in## Loaded memory entries. Seeskills/repo-skillfor format and authoring guide. - Read both indexes (if they exist):
~/.kon/public/memory/MEMORY.md~/.kon/projects/<repo-name>/memory/MEMORY.md
- Parse index lines
- [Title](file.md) — descriptionfrom each; tag each row with scope[public]or[repo:<name>]. - Rank by relevance to the current task (keywords / topic overlap with description).
- Repo wins ties — same topic in both indexes → prefer the repo entry.
- Load top 10 entry files total (not 10 per index).
- Print
## Loaded memory entrieslisting scope + title for each entry used.
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.
- yesterday First seen · 84 lines · 0 tokens per session scan A 48b12eeae4e2
memory-loading is a skill published in the GitHub repository dentiny/kon (3 stars, last pushed 1mo ago), licensed MIT. It adds 58 tokens to every session and 999 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…