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/sergeyklay/.agents/log-changesnpx skills add sergeyklay/.agents --skill log-changesgit clone --depth 1 https://github.com/sergeyklay/.agentsWrote 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/sergeyklay/.agents/log-changes)<a href="https://agentmods.dev/skills/sergeyklay/.agents/log-changes"><img src="https://agentmods.dev/badge/skills/sergeyklay/.agents/log-changes.svg" alt="Measured on agentmods" 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 | $0.00096 | $0.06594 |
| Opus 5 | $0.00048 | $0.03297 |
| Sonnet 5 | $0.00019 | $0.01319 |
| Haiku 4.5 | $0.00010 | $0.00659 |
Grade A, and why
log-changes 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 yesterday.
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 — 328 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Changelog Maintenance
The changelog records notable changes to the distributed software. Every entry must answer: "Does this change affect someone who uses, upgrades, deploys, or integrates with the project?" If not, omit it.
Format authority: Keep a Changelog 1.1.0 and Semantic Versioning 2.0.0.
Project constants
Detect the following constants from the project itself - do not ask the user unless detection fails. Inspect, in order: existing CHANGELOG.md entries, project documentation (README.md, CONTRIBUTING.md, docs/), and recent git commit messages.
-
GitHub repository slug (
OWNER/REPO). Read existing comparison links at the bottom of CHANGELOG.md (preferred). Fall back togit remote get-url originand parse the slug from the URL. -
Issue tracker. Look for tracker references in existing CHANGELOG entries, PR templates, contributing docs, and recent commit messages:
- GitHub Issues - links of the form
https://github.com/OWNER/REPO/issues/NNN, or closing keywordscloses/fixes/resolves #NNNin commit/PR bodies. - Jira - task keys of the form
[A-Z]+-[0-9]+(e.g.ABC-123) or links to*.atlassian.net/browse/.... - Linear - keys like
ENG-123or links tohttps://linear.app/.... - Other trackers - distinct ID schemes or links in commit/PR bodies.
For non-GitHub trackers, also detect the tracker base URL and the project key prefix (e.g.
BP,ABC,ENG). - GitHub Issues - links of the form
-
Subsystem labels. Read existing CHANGELOG entries to learn which subsystem prefixes the project already uses (e.g.
API:,CLI:,Auth:). If none exist, propose labels that match the project's top-level directory or package layout. The human reviewer will correct any that are wrong. -
Entry unit. Count tracker references per bullet in the newest released section to learn whether the project groups distinct but related user-facing changes by epic, milestone, or roadmap checkpoint. This convention controls whether separate logical changes may share a bullet; it never splits one logical change by PR, commit, or implementation task.
-
Bullet order within a category. Keep a Changelog fixes the order of versions and categories but says nothing about bullets inside a category, so every project has its own convention. Recover it from git rather than guessing: for three or four bullets in the newest released section, run
git log --format='%h %ai %s' -S'<key unique to that bullet>' -- CHANGELOG.md | tail -1to find the commit that added it. Descending timestamps down the section mean newest-first (prepend); ascending mean oldest-first (append). Default to newest-first when the file is new or the signal is mixed - it matches the reverse-chronological rule the file already applies to versions. -
Audience. Decide who reads the file: operators and consumers of a distributed artifact (published package, self-hosted service, OSS release), or an internal team plus non-technical stakeholders such as a client, manager, or leadership. This governs whether deploy mechanics belong in an entry at all - see Step 4. Detect it from the project's distribution setup (release workflow, package manifest, install docs); ask once if that is inconclusive.
-
Product surface. Detect whether the project ships an application, CLI, service, library, or framework from its install and usage documentation. In applications, CLIs, and services, implementation types, functions, modules, protocol states, and internal error taxonomies are not user-facing merely because they have names. In libraries and frameworks, include a symbol only when it is part of the supported public API.
What ships with it
2 files 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.
- yesterday Changed · +41 lines · +1 tokens per session 06a6954f20c9
- 5d ago First seen · 287 lines · 95 tokens per session scan A 0d8a00814998
log-changes is a skill published in the GitHub repository sergeyklay/.agents (5 stars, last pushed today), licensed Apache-2.0. It adds 96 tokens to every session and 6,594 once invoked, about $0.0005 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
release-check
Pre-release checklist for Forge — verify, version bump, changelog, docs, docker build, smoke test. Run before tagging v.
landing
End-of-session ritual. Close finished beads, file new ones for remaining work, run verify, commit in logical groupings, push. Work is not complete until git push succeeds.
proofslip-release-proof
Verify a ProofSlip release proof or add provider-backed release proof to a GitHub Actions deployment workflow. Use when a coding agent or CI job claims it shipped, when another agent must decide whether to trust a release claim, or when a repository needs the minimal ProofSlip GitHub OIDC setup. Do not use it to treat…
agent-release-swarm
Agent skill for release-swarm - invoke with $agent-release-swarm.
agent-release-manager
Agent skill for release-manager - invoke with $agent-release-manager.
agent-github-modes
Agent skill for github-modes - invoke with $agent-github-modes.