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 sammcj/agentic-coding --skill ai-changeloggit clone --depth 1 https://github.com/sammcj/agentic-codingWrote 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/sammcj/agentic-coding/ai-changelog)<a href="https://agentmods.dev/skills/sammcj/agentic-coding/ai-changelog"><img src="https://agentmods.dev/badge/skills/sammcj/agentic-coding/ai-changelog.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00081 | $0.01905 |
| Opus 5 | $0.00041 | $0.00953 |
| Sonnet 5 | $0.00016 | $0.00381 |
| Haiku 4.5 | $0.00008 | $0.00191 |
Grade A, and why
ai-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 7d 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI-Driven Changelog
Set up a changelog system AI agents maintain during development. Two shapes, chosen by project type:
- Software projects: agents write entries under
## [Unreleased]; automation stamps version numbers at release time (CalVer or SemVer). No agent ever writes version numbers; the build process handles that. - Non-code projects (no build system): agents add terse bullets under a date heading (
## YYYY-MM-DD) by hand. No script, no version numbers. Seereferences/dated.md.
Setup workflow
-
Detect the build system: Check for Makefile, Justfile, package.json, Cargo.toml, pyproject.toml, go.mod. Note which config files contain a
"version"field. -
Check whether this is a software project at all. If there's no build system from step 1 and no version convention (a docs repo, writing or notes vault, content site, research or config collection), use Dated mode: date-based changelog headings, no version script, no build integration. Skip straight to
references/dated.mdand ignore the script/build steps below. Otherwise, detect the versioning scheme by inspecting (highest confidence first):CHANGELOG.mdheading style:## [YYYY.M.N]headings → CalVer;## [X.Y.Z]headings or prose mentioning "SemVer" → SemVer- Git tags from
git tag --list | head:vX.Y.Z→ SemVer;YYYY.M.N→ CalVer VERSIONfile with content matching^[0-9]+\.[0-9]+\.[0-9]+→ SemVer- Manifest version field (
package.json,Cargo.toml,pyproject.toml) matchingX.Y.Z→ SemVer
If signals are absent or contradictory, ask the user. Suggest SemVer for projects with established version history (existing tags, manifest versions, prior changelog entries) and CalVer for greenfield projects where automatic versioning is preferable. Each scheme has trade-offs documented in its reference file.
-
Read the scheme reference that matches the chosen scheme. It contains the build integration recipes, CLAUDE.md snippet, GitHub Actions pattern, and scheme-specific gotchas:
- CalVer →
references/calver.md - SemVer →
references/semver.md - Dated (non-code projects) →
references/dated.md
- CalVer →
What ships with it
5 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.
- 7d ago First seen · 85 lines · 81 tokens per session scan A c51732c2ffa1
ai-changelog is a skill published in the GitHub repository sammcj/agentic-coding (160 stars, last pushed today), licensed Apache-2.0. It adds 81 tokens to every session and 1,905 once invoked, about $0.0004 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
agentpush
Bridge an imported agentpush MCP server (Telegram, WhatsApp, etc.) with live agentproto sessions via the daemon's transmitter subsystem: transmitmessage sends outbound and binds the recipient to a session, inboundwatcherstart polls agentpush for new messages, and inboundendpointcreate/POST /inbound(/:slug) route…
bureau
Drive Bureau — the browser stack's installable capability server: a stealth Firefox (Camofox) + daemon that exposes browser tools as MCP-over-HTTP on :8830, plus a CLI for saved browser identities (sessions), social capture / search, adapter health probes, declarative workflows, and Guilde connection. Use when working…
agentproto-apps
Operate and build Agentproto apps — AIP-42 app bundles (defineApp().emit(dir), APP.md + agents/ + workflows/ + ui/) and their daemon lifecycle (appinstall, appapply, apprun, appstatus, appstop) plus the app-scoped durable data plane (appdataread/write/list/migrate). Covers serving one with a UI (agentproto app serve →…
agentproto-llm-endpoint
Use the local LLM endpoint proxy (localhost:18090) — or your own gated public deployment — to route Claude Code / Claude SDK through custom providers (OpenRouter, Moonshot, Groq, ZAI) via an Anthropic-compatible gateway. Covers adapter-specific behaviors, auth patterns, proxy codenames, and common failure modes.
Drive the user's real authenticated LinkedIn — via camofox stealth (preferred, distinct session + WhatsApp re-auth alert) or the chrome-devtools chain (fallback). API-first: read feed/profiles/activity, who-reacted/commented, connections, people + content search (who posted about a topic, with resolved permalinks)…
Drive the user's authenticated Instagram via the browser skill: profiles, media likers/comments, followers/following, search via the web private API; gated like/comment/follow/DM. Use for Instagram research/outreach as the user.