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 commands/agent-marketplace-kit/cli/syncgit clone --depth 1 https://github.com/AGent-marketplace-KIT/cliWhat 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 | $0.00013 | $0.00498 |
| Opus 5 | $0.00006 | $0.00249 |
| Sonnet 5 | $0.00003 | $0.00100 |
| Haiku 4.5 | $0.00001 | $0.00050 |
Grade A, and why
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 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.
What it actually says
agmk sync
Reconcile marketplace.json and the generated docs with the plugins actually on
disk. The source of truth is each plugin's .claude-plugin/plugin.json.
agmk sync
add and bump run sync internally, so you call it directly mainly after
hand-editing a plugin manifest or moving files around.
What it does
Scanning every directory under pluginRoot that has a
.claude-plugin/plugin.json, sync:
- adds catalog entries for plugin directories not yet listed (with a
marketplace-root-relative
"./…"source); - updates any
version,description,author,keywords, orhomepagethat drifted from the manifest into the catalog; - flags orphans — catalog entries whose local source directory no longer
exists (remote
github/url/git-subdirsources are left untouched); - sorts plugins by name and writes
marketplace.json; - refreshes any already-built Codex/Cursor registries;
- regenerates the plugin table in
README.mdand the plugin list inAGENTS.md, between the<!-- agmk:plugins:start -->/<!-- agmk:plugins:end -->markers.
{% hint style="info" %}
Drop the same agmk:plugins markers into any other Markdown file and sync
maintains the plugin list there too.
{% endhint %}
Output & exit code
Prints the list of changes (or "Catalog already in sync."). If any orphan
is detected, it prints a ! line telling you to remove the entry or restore the
plugin, and exits non-zero.
Example
# after editing plugins/tdd-coach/.claude-plugin/plugin.json
agmk sync
# ~ updated description of "tdd-coach"
# ~ refreshed plugin list in README.md
See also
agmk add·agmk bump·agmk validate- Reference: marketplace.json
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 First seen · 58 lines · 13 tokens per session scan A 741209bf9601
sync is a command published in the GitHub repository AGent-marketplace-KIT/cli (6 stars, last pushed 1mo ago), licensed MIT. It adds 13 tokens to every session and 498 once invoked, about $0.0001 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 commands, from other repositories
annotate-plan
Draft a plan and open it in the annotator UI for user review.
docs
Command "docs" from tilework-tech/nori-skillsets, covering noridoc: commands, how it fits into the larger codebase, core implementation and things to know.
build-loop
Planning Mode — scaffold a target project's l00prite/ folder (CLAUDE.md blueprint, AGENTS.md, .l00prite memory, universal loop prompts), root pointer files, vendor adapters, and a skeleton repo from a project idea. Stops after scaffolding; --execute hands off to Execution Mode only through execute-loop's confirmed…
event-loop
You are processing one pending l00prite event. Events are protocol objects, not vendor-specific automation commands.
respond-to-review
You are responding to one review-related l00prite event. Treat .l00prite/ as the shared source of truth across agents.
resume-loop
You are resuming a l00prite-managed project. Treat .l00prite/ as the shared source of truth across Claude, Codex, GPT, Gemini, and future agents.