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 skills add svy04/ballast --skill knowledge-basegit clone --depth 1 https://github.com/svy04/ballastWrote 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/skills/svy04/ballast/knowledge-base)<a href="https://agentmods.dev/skills/svy04/ballast/knowledge-base"><img src="https://agentmods.dev/badge/skills/svy04/ballast/knowledge-base/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/skills/svy04/ballast/knowledge-base"><img src="https://agentmods.dev/badge/skills/svy04/ballast/knowledge-base.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00047 | $0.00964 |
| Opus 5.5 | $0.00019 | $0.00386 |
| Sonnet 5.5 | $0.00009 | $0.00193 |
| Haiku 4.5 | $0.00005 | $0.00096 |
Grade A, and why
knowledge-base 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 1mo 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 — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowledge base
A session that researches well and forgets is a session that will research again. This skill adds the layer that makes learning cumulative: verified findings land in memory/knowledge/<topic>.md, and every new question checks there first.
Lookup before research — always
Before investigating any topic, read memory/knowledge/ for an entry that answers it. If a current entry exists, use it and cite it — re-researching a settled question wastes the work that settled it. If the entry is stale or only partially covers the question, research the gap, not the whole topic.
Drafts stay drafts
Findings made mid-task (a benchmark result, a library's actual behavior, an API's undocumented limit) are drafts. Label each with the verify-gate labels — confirmed / observed / assumed / hearsay / unknown — wherever the draft lives: session log, working notes, a comment in the task file. Drafts may inform the current task but must not be quoted later as established knowledge.
Promotion — the gate is the door
Only claims that pass the verify gate enter memory/knowledge/. One file per topic, entries shaped like:
## Connection pool exhausts under burst load — verified 2026-08-14
- Claim: the default pool (10) exhausts at ~40 concurrent requests; queue wait exceeds 2s.
- Sources: load test output (tests/load/pool.txt), library docs on pool sizing (both opened).
- Sample: 3 runs, same config. Limits: measured on the staging tier only; production tier untested.
Every entry carries: the claim, named primary sources, sample size, stated limits, and a verification date. An entry missing any of these is still a draft, whatever file it sits in.
The first time you create memory/knowledge/, add a row for it to the File map in memory/00-INDEX.md before writing the first entry — a folder missing from the map is a folder the next session never opens.
Check what the finding contradicts before filing it. Read the existing entries this finding touches. If it contradicts one, do not shelve the two beside each other: surface the conflict first, then either supersede the old entry — with a link back to it from the new one — or hold the new entry and register the question in OPEN-QUESTIONS.md. If the two rest on different layers of evidence (different scope, subject, or measurement), placing them side by side is as far as it goes; they do not get merged into one claim.
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.
- 1mo ago First seen · 57 lines · 47 tokens per session scan A 33380b3c24f3
knowledge-base is a skill published in the GitHub repository svy04/ballast (72 stars, last pushed 2mo ago), licensed MIT. It adds 47 tokens to every session and 964 once invoked, about $0.0002 per session on Opus 5.5. 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 skills, from other repositories
alive:save
The human wants to checkpoint. Or: the stash has grown heavy — 5+ items, 30+ minutes, a natural pause in the work. The squirrel doesn't decide when to save. It surfaces the need and lets the human pull the trigger. Runs the full save protocol: confirms stash, writes log, updates state, generates projections…
alive:capture-context
Use when external content arrives in the session — emails, transcripts, screenshots, documents, files, or in-session research worth keeping. Also use when there's nothing obvious to capture — the skill checks 03Inbox/ for unrouted files and enters inbox scan mode. Stores raw content, routes to bundles, extracts tasks…
alive:session-history
Revive sessions (quick or heavy), browse, and search — 'what happened recently?', 'find the session where we discussed X', 'revive yesterday's session'. For single-session recall and multi-session browsing. If the human needs to merge multiple sessions into one working context or detect conflicts between parallel…
alive:load-context
The human mentions a walnut to work on, asks about a specific venture/experiment/project, or wants to check status — not just explicit 'load X'. Load the brief pack (3 files), resolve the people involved, check the active bundle — then surface one observation and ask what to work on. Context loads in tiers: walnut and…
alive:mine-for-context
Deep context extraction from source material. Creates reference bundles, builds extraction plans, tracks what's been extracted, and discovers new targets — people, subjects, patterns, connections. The archaeologist that turns raw sources into structured knowledge. Can be invoked by alive:session-history for targeted…
alive:my-context-graph
Render an interactive map of your world. Generates the world index from all walnut and bundle frontmatter, then produces a force-directed graph showing connections between walnuts, people, bundles, and tags. Opens in the browser.