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 recallgit 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/recall)<a href="https://agentmods.dev/skills/svy04/ballast/recall"><img src="https://agentmods.dev/badge/skills/svy04/ballast/recall.svg" alt="Measured on agentmods" 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 | $0.00044 | $0.00835 |
| Opus 5 | $0.00022 | $0.00417 |
| Sonnet 5 | $0.00009 | $0.00167 |
| Haiku 4.5 | $0.00004 | $0.00084 |
Grade A, and why
recall 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 8d 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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Recall
The failure this prevents is not missing knowledge. It is knowledge that was written down, indexed, and then not opened.
That failure has a shape: the answer existed in the project's own files, the assistant answered from general habit instead, and nobody noticed until the user said "isn't that already written down somewhere?" — because it was. Delivery hooks do not close this. A hook fires on keywords in the message; it cannot know that a settled note three folders away decides the question.
When it runs
Two moments, no exceptions:
- The first substantive reply of a session. Not the greeting — the first answer that carries work.
- A subject shift. The thread moves to something it was not about: a different deliverable, a different field, a different kind of question. If you are unsure whether the subject shifted, it shifted.
Inside one continuing subject, later replies do not repeat the sweep. It has already run.
What to sweep — five places, all of them
memory/00-INDEX.md— the map of what exists at all. Read this first; it names the layers the rest of the sweep will visit.memory/knowledge/— findings that passed the gate.DECISIONS.md— what was settled, and what superseded what.- The rule catalog (
.claude/ballast.rules.json, plus the user-level one) — standing corrections that apply whether or not a keyword matched this message. skills/andmemory/goal/— procedures already forged, and the skeleton of any goal in flight.
Scan at the index level — titles, headings, entry names. Then open what looks relevant and actually read it before answering.
Do not stop at the first hit
A sweep that ends when something turns up is the same failure wearing a better hat.
- Finding one does not end the sweep. Visit all five places even after an answer appears in the second.
- The layers hold different things. A rule is not a verified fact; a verified fact is not a settled decision; a settled decision is not a forged procedure. One layer answering the question does not mean the others have nothing to add — often they qualify or contradict it.
- Open widely, not carefully. Read the candidates together rather than one at a time, and delegate a broad pass when the surface is large. Opening fewer files is not economy; it is the omission this skill exists to prevent.
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.
- 8d ago First seen · 54 lines · 44 tokens per session scan A 51ff64feb2a0
recall is a skill published in the GitHub repository svy04/ballast (71 stars, last pushed 13d ago), licensed MIT. It adds 44 tokens to every session and 835 once invoked, about $0.0002 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 2026-08-30.
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: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: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: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.