All Claude Code modsconnectwithub/loopkeeper
lk-token-budget
Stop runaway token spend: budgets per session and per subagent with warn, deny-agents or stop actions, plus a prompt-cache countdown, cold-cache pings and keep-warm.
lk-token-budget reaches L2 (writes or runs) and passed claude plugin validate on Claude Code 2.1.312.
- Events hooked: 12
- API methods referenced: 19
- Access from classified calls: L2, writes or runs: drives Claude, reads files, reads env vars, reads mod state, persists state, draws
- Reads mod state: the scan does not record whose state, so this does not show that it reads another mod’s values
- Static scan, not a runtime test
Install hint
/plugin install lk-token-budget --marketplace connectwithub/loopkeeperFrom the repository's own marketplace. Confirm the plugin name in the repository's README before installing. Mods need Claude Code 2.1.287 or later.
Access
L2 Writes or runsdrives Claude, reads files, reads env vars, reads mod state, persists state, drawsValidation
Passed on Claude Code 2.1.312Listed under Prompt cache mods for Claude Code, Context and usage meters for Claude Code.
View the source on GitHub Back to the list
What the scan recorded
- Observed events
- classic hooks, compaction, every prompt, subagent spawns, every tool call
- Hooks
session.start, classic.SessionStart {source=clear|resume|fork}, session.compact, prompt.submit, session.end, turn.step, turn.start, turn.complete, agent.spawn, tool.call, command.run {command=budget}, ui.render {component=AbovePrompt}, ui.render {component=Pane, requestId=token-budget}- API calls
$.clock.every, $.clock.now, $.command.register, $.env.get, $.fs.exists, $.fs.read, $.model.fork, $.session.id, $.session.usage, $.state.get, $.state.set, $.store.get, $.store.set, $.ui.close, $.ui.invalidate, $.ui.log, $.ui.open, $.ui.resolve, $.ui.toast- Validation on 2.1.312
- Passed
- Scanned revision
ff413ab21be1- Marketplace passed
- loopkeeper
These are the hooks and API calls reported by claude plugin validate. Access levels describe the widest reach of those calls and are not safety ratings. Validation is a static check, not a runtime test, and a passing result does not prove that a mod works or is safe. What a mod can access and how the scan works.
More from connectwithub/loopkeeper
- lk-agent-mapSee and cap your subagents: cards with cost, tokens, context and progress, spawn caps with an override you ap…
- lk-attentionLook away while Claude works: a GNOME notification, a sound, a toast and a band when it needs you or finishes…
- lk-handoff-guardNever lose a session to the context limit: it watches context fill, writes a handoff into .claude/brain/ in t…
- lk-pr-watchKnow when CI breaks without leaving the session: PR and check state above the prompt, a /pr pane, toasts on c…
- lk-secret-guardA seatbelt for secrets: stops credential-shaped strings and your private deny-list of words before Claude wri…
- lk-where-am-iKnow where the session is at a glance: repo, branch, worktree, uncommitted and unpushed work, usage windows a…
- loopkeeperKeep Claude Code on a plan: stages, approval gates and a plan-scoped file budget enforced by hooks, plus mode…
Badge for your README
Authors can show the latest scan result in their README. Each badge links to this page.
[](https://mods.aidojo.si/mods/connectwithub--loopkeeper--lk-token-budget/)
[](https://mods.aidojo.si/mods/connectwithub--loopkeeper--lk-token-budget/)