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/osteele/agent-lore/lore-sweepnpx skills add osteele/agent-lore --skill lore-sweepgit clone --depth 1 https://github.com/osteele/agent-loreWhat 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 | $0.00110 | $0.01026 |
| Opus 5 | $0.00055 | $0.00513 |
| Sonnet 5 | $0.00022 | $0.00205 |
| Haiku 4.5 | $0.00011 | $0.00103 |
Grade A, and why
lore-sweep 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 2d 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 — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sweeping lore's recorded gaps
A zero-result search is a topic gap stated in an agent's own words, at the moment it cost
somebody something. lore misses turns those into a queue. This skill works the queue.
The queue is self-cleaning: every candidate is replayed against the current repo before it is
listed, so a gap someone has already filled drops off and is counted under resolved. You
never have to prune it by hand, and you should never "close" a row by editing the log.
1. Read the queue
lore misses --json # for working through programmatically
lore misses # human-readable
lore misses --since 30d --min-count 2 # only gaps that cost more than one session
Each row carries the query verbatim, how many times it was hit, which sessions hit it, the directory each was working in, and a suggested path.
2. Triage before you write
Not every row is a missing page. Sort each into one of four:
- Answerable now. The fact is recoverable from this machine — a man page,
--help, the source, a quick experiment. Most rows. Go to step 3. - Answerable only from the session that hit it. The session worked the problem out and the
resolution is in its transcript. Use the
ctx-agent-history-searchskill, scoped to the cwd the row gives you. This is where the highest-value facts live, because a gap someone had to solve the hard way is exactly the gap worth recording. - Not a gap — a category error. The session was addressing lore as if it were something
else. A row like
lore_read plans/backlog/2026-07-26-….mdis a session looking for its own project's lab notebook; lore holds what outlives a single project. Do not write a page. Leave the row; it costs nothing and it is evidence about how sessions misread the tool. - Still unknown. Nobody knows the answer yet. Leave it. A row that sits is honest.
Say which bucket each row went into. Do not silently skip a row.
3. Search lore before you write
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.
- 2d ago First seen · 81 lines · 110 tokens per session scan A 6e69a1c6feea
lore-sweep is a skill published in the GitHub repository osteele/agent-lore (0 stars, last pushed 7d ago), licensed MIT. It adds 110 tokens to every session and 1,026 once invoked, about $0.0006 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
phase-orchestrator
Orchestrates phase-based implementation plans using the host's todo tracker and subagents for implementation, review, validation, default phase commits, plan finalization, and a final end-to-end review. Delegates work through the companion phase-implementer, phase-reviewer, and phase-final-reviewer skills and consumes…
sceneproof
Visual reasoning discipline for verifying UI and Three.js work with the SceneProof CLI. Use whenever a change can alter rendered output — components, styles, layout, SVG, canvas, scenes, cameras, lights, materials, animation — or whenever about to claim that something "looks right", "is visible", or "renders…
write-phase-plan
Writes or restructures an implementation plan in the format the phase-orchestrator skill consumes: dependency-ordered phases with explicit scope, acceptance criteria, validation commands, and status/log conventions. Use when the user asks to write a plan for phase orchestration, prepare a plan or PRD for delegated…
kb
Operate a hraness/kb local-first Markdown knowledge base for coding-agent memory. Use when a user asks to search or query a KB or Obsidian vault; load repository context, plans, decisions, concepts, backlinks, semantic search, or Git provenance; save, clip, scrape, or archive a URL, article, social thread, signed-in…
setup
Set up memex with provisioned Postgres (RDS + pgvector), AGENTS.md injection, first import.
cold-start
Day-one data bootstrapping for a new brain. Sequences the highest-leverage data sources to go from empty brain to useful brain in one session. Uses the agent host's managed connectors for safe credential handling — the agent never holds raw API keys. Covers Gmail import, calendar sync, contacts seeding, X/Twitter…