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 deepklarity/harness-kit --skill hk-shift-changeloggit clone --depth 1 https://github.com/deepklarity/harness-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/deepklarity/harness-kit/hk-shift-changelog)<a href="https://agentmods.dev/skills/deepklarity/harness-kit/hk-shift-changelog"><img src="https://agentmods.dev/badge/skills/deepklarity/harness-kit/hk-shift-changelog/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/deepklarity/harness-kit/hk-shift-changelog"><img src="https://agentmods.dev/badge/skills/deepklarity/harness-kit/hk-shift-changelog.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 | $0.00111 | $0.00820 |
| Opus 5 | $0.00056 | $0.00410 |
| Sonnet 5 | $0.00022 | $0.00164 |
| Haiku 4.5 | $0.00011 | $0.00082 |
Grade A, and why
hk-shift-changelog 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 10d 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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Shift-handover changelog (project-agnostic)
One living HTML page, rewritten (never appended) at every meaningful transition, redeployed to the SAME artifact URL every time. The reader is a teammate back from sleep or a weekend: they want the state of the world, not a log.
Where it lives
- File: one stable path in the repo (e.g.
docs/<area>/changelog.html), committed, so any future session can find and redeploy it. - URL: record the artifact URL in the project's durable operator notes
(state/RESUME file) the first time you publish. Every later publish
passes that
url— never mint a second artifact for the same duty. - Favicon: pick one emoji the first time; never change it (users find the tab by icon).
Structure (top-down by how fast the reader wants out)
- Header — project + "shift handover", a stamp saying what period it covers (relative, not raw timestamps).
- TL;DR box — 3-5 sentences: what changed, what's running, whether anything is blocked on the reader. If they read nothing else, this suffices.
- What happened, in order — timeline entries: a short label (wave close / incident / directive / decision), a bold headline, 1-3 sentences, and a status pill (DONE / RUNNING / VERIFIED / WAITING). Every user directive gets its own entry so they can verify they were understood.
- Scoreboard / metrics — whatever the project tracks (bucket scores, suites, costs), as a table with movement + evidence, never bare deltas.
- Waiting on you — a numbered list of decisions only the human can make. This section earns the page its keep; keep it brutally current.
- Footer — "living document, rewritten each transition" + where the ground truth lives.
Design rules
Load the artifact-design skill before first authoring. Beyond it:
- Utilitarian treatment: real hierarchy, quiet type, one accent; both
themes via CSS tokens (
prefers-color-scheme+data-themeoverrides). - Bind background AND text color to the SAME token set on
html, bodywith!important. The artifact shell may force its own body background; if your ink tokens go dark while its ground stays light, the page is unreadable (seen live). Background and ink must move together or not at all. - Self-contained (CSP): no external fonts/scripts; system font stacks.
- Tables wrapped in
overflow-x:auto; digits gettabular-nums. - Plain language throughout; no ids or jargon the reader must decode; name
things by what they mean, cite ids only as secondary
monodetail.
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.
- 10d ago First seen · 64 lines · 111 tokens per session scan A 9666b3f72771
hk-shift-changelog is a skill published in the GitHub repository deepklarity/harness-kit (97 stars, last pushed 1mo ago), licensed MIT. It adds 111 tokens to every session and 820 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 2026-08-30.
Other skills, from other repositories
product-plan
Interactive planning session - create PRD and PROGRESS files for a feature.
status
Show current status of belmont tasks and milestones.
parallel-feature-development
Coordinate parallel feature development with file ownership strategies, conflict avoidance rules, and integration patterns for multi-agent implementation. Use this skill when decomposing a large feature into independent work streams, when two or more agents need to implement different layers of the same system…
track-management
Use this skill when creating, managing, or working with Conductor tracks - the logical work units for features, bugs, and refactors. Applies to spec.md, plan.md, and track lifecycle operations.
neuron-structured-output
Design and implement structured output classes for Neuron AI agents using SchemaProperty attributes and validation rules. Use this skill when the user mentions structured output, JSON schema extraction, data validation, output classes, DTOs for AI responses, extracting structured data from LLM, or configuring property…
organize-task-files
Lists, pairs, deduplicates, and moves task files across the Coworker task state machine (0draft → 6git-pushed). Use when asked to organize, clean up, or audit task files.