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 iowarp/clio-kit --skill session-recordgit clone --depth 1 https://github.com/iowarp/clio-kitWrote 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/iowarp/clio-kit/session-record)<a href="https://agentmods.dev/skills/iowarp/clio-kit/session-record"><img src="https://agentmods.dev/badge/skills/iowarp/clio-kit/session-record/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/iowarp/clio-kit/session-record"><img src="https://agentmods.dev/badge/skills/iowarp/clio-kit/session-record.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 | $0.00048 | $0.00742 |
| Opus 5.5 | $0.00019 | $0.00297 |
| Sonnet 5.5 | $0.00010 | $0.00148 |
| Haiku 4.5 | $0.00005 | $0.00074 |
Grade A, and why
session-record 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 9d 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Session Provenance
ChronoLog stores the conversation — user messages and the responses to them — in a named chronicle and story. It is a provenance trail, not a profiler.
This is worth being precise about, because the server sits in a bundle with the
I/O profiler and the two get confused. ChronoLog has nothing to say about how
long a job ran or what it did to the filesystem. For that, see
job-diagnose.
The handle is the whole thing
clio-chronolog:start_chronolog connects, creates the chronicle, and acquires a
story handle. Everything after depends on that handle existing:
clio-chronolog:start_chronolog— connect, name the chronicle and storyclio-chronolog:record_interaction— log a message and its responseclio-chronolog:stop_chronolog— release the handle and disconnect
record_interaction without a live handle fails. Recording is not automatic
either: nothing is captured unless the call is made for it, so a session with one
start_chronolog and no record_interaction produces an empty story.
Always stop. The handle is a held resource on the ChronoLog side. Leaving it open after the work is done leaks it, and the story may not be readable until it is released.
Reading it back
clio-chronolog:retrieve_interaction takes the chronicle and story names, with
optional time filtering. This is a separate operation from recording — reading an
old story needs no start_chronolog for it.
Name chronicles and stories for how they will be searched later. A story called
session is unfindable a month afterwards; one named for the experiment and date
is not.
When this is worth doing
Recording every session is pure overhead. It earns its place when the reasoning matters as much as the result: a parameter sweep where the choices need justifying later, a debugging session someone else will pick up, or work that feeds a paper and has to be reconstructable.
What not to do
- Do not use ChronoLog to investigate job performance.
- Do not call
record_interactionbeforestart_chronolog. - Do not leave the handle open at the end of a session.
- Do not assume anything was captured — recording is explicit, per interaction.
- Do not name a chronicle something you could not search for later.
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.
- 9d ago First seen · 73 lines · 48 tokens per session scan A 4c4326947fea
session-record is a skill published in the GitHub repository iowarp/clio-kit (30 stars, last pushed 8d ago), licensed BSD-3-Clause. It adds 48 tokens to every session and 742 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 2026-10-08.
Other skills, from other repositories
bioprobench
Score an LLM's biological-protocol reasoning on the BioProBench benchmark: protocol QA, step ordering, error detection, protocol generation, and LLM-judged error reasoning; or generate the responses.
evo-memory
Manages persistent research memory across ideation and experimentation cycles. Maintains two stores: Ideation Memory MI (feasible/unsuccessful directions) and Experimentation Memory ME (reusable strategies for data processing, model training, architecture, debugging). Three evolution mechanisms: IDE (after…
raindrop-mcp
Connect Raindrop.io to your AI assistant with a simple MCP server. Use it to organize, search, and manage bookmarks with natural language.
onescience-orchestrator
A task coordinator for OneScience that turns a user’s goal into a tracked plan and assigns parts of the work to suitable specialist tools.
sota-management
4-level SOTA performance tracking for VibeCodeHPC. Use when checking or updating best performance records.
authoring-gradle-builds
Authoring and modifying Gradle build logic, including settings, plugins, dependencies, tasks, conventions, and upgrades. Activate when build files or plugins must change; use using-gradle for inspection or execution without build-definition edits.