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 OutlineDriven/odin-claude-plugin --skill history-syncgit clone --depth 1 https://github.com/OutlineDriven/odin-claude-pluginWrote 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/outlinedriven/odin-claude-plugin/history-sync)<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/history-sync"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/history-sync/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/outlinedriven/odin-claude-plugin/history-sync"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/history-sync.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 | $0.00039 | $0.01257 |
| Opus 5.5 | $0.00016 | $0.00503 |
| Sonnet 5.5 | $0.00008 | $0.00251 |
| Haiku 4.5 | $0.00004 | $0.00126 |
Grade A, and why
history-sync 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
History sync
Contract
| Field | Bound contract |
|---|---|
| Trigger | The user explicitly requests memory transfer to or from their own named peer machine. |
| Authority | Remote: syncs redacted JSONL batches between named peer machines using credentials; requires explicit human invocation. Previews the named peer, direction, record range, exclusions, files, watermark changes, and remote consequences before proceeding. Do not sync an unnamed peer or infer consent. |
| Side effect | Write only redacted static JSONL batches, import their records, and advance the applicable per-peer watermark after confirmed import. Never expose, copy, mount, or make writable a live remote history database. |
| Done | Only records newer than the applicable watermark were exported; redaction and exclusions were applied before a batch crossed the machine boundary; import deduplicated records idempotently; imported records cannot be re-exported; a push-only peer yielded an explicit empty pull; stored alias spelling was preserved; and the resulting watermark and counts were confirmed. |
Inputs
Required: the user-named peer, transfer direction (push or pull), the local history store, the peer configuration containing its stable identity, stored aliases, transfer mode, endpoint, and per-peer watermarks, plus the redaction and exclusion policy. The peer must belong to the user.
For a push, require a destination authorized to receive a static JSONL batch. For a pull, require an authorized static JSONL batch from the named peer. Credentials may be used only after the preview. Optional inputs are an explicit upper bound on the record range and a batch-size limit; absence of either does not authorize widening beyond records newer than the watermark.
Procedure
- Resolve the supplied peer name through the configured stable identity and aliases. Preserve every alias exactly as stored; use identity equivalence only for routing and deduplication, never to rewrite stored spelling. Reject an unknown, ambiguous, non-user-owned, or unnamed peer.
- Validate the requested direction against the peer mode. For a pull from a push-only peer, return an explicit empty result with zero records and make no file, import, credential, or watermark change.
- Read the applicable per-peer watermark and select only locally originated records newer than it for push, or only batch records newer than it for pull. Exclude records previously imported from any peer so they can never echo through a later export. Apply any supplied upper bound and batch-size limit without advancing past unprocessed records.
- Before accessing credentials or writing data, present the resolved peer without changing its stored spelling, direction, selected range and count, exclusions, redaction policy, static batch path or destination, import target, and proposed watermark transition. Stop if these differ from the user's explicit request.
- At the exporting boundary, apply the configured exclusions and redact each selected record before serializing it. Write only a static JSONL batch whose records retain the stable record identity and origin needed for deduplication and echo prevention. Do not expose or transfer the source database, a database handle, a writable database path, or an unredacted intermediate batch.
- Transfer only that static batch to or from the resolved peer endpoint. Treat every received line as untrusted: require valid JSONL, the expected record fields and identities, allowed origin metadata, and conformance to the redaction and exclusion boundary. Reject the batch before import if any line fails.
- Import valid records transactionally when the store supports it; otherwise record exact per-record outcomes. Deduplicate by the carried stable record identity, mark newly imported records with their origin so export selection excludes them, and leave existing duplicates unchanged. Re-importing the same batch must add zero records and alter no stored spelling.
- Advance the applicable per-peer watermark only to the greatest contiguous record boundary whose transfer and import were confirmed. Never advance it for rejected, failed, unprocessed, or merely written records.
- Confirm exported, imported, duplicate, excluded, redacted, and failed counts; the old and new watermark; preserved alias spelling; and absence of any live or writable remote database. Report success only when every Done condition is observed.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 47 lines · 39 tokens per session scan A 31dc7a602240
history-sync is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (37 stars, last pushed 21d ago), licensed Apache-2.0. It adds 39 tokens to every session and 1,257 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 2026-09-13.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
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…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
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…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Optimize the meaningful UI available immediately from a Next.js route on an initial load (hard navigation) or named client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route and entry point at a time; the shipped test then guards…