Session Budget
Assess whether compacting or clearing the current session is worth it right now, and whether a reinit-ready handoff that survives /clear would carry real value. Not a threshold nag — it X-rays context composition (how much is stale vs. still load-bearing), which only the agent can see. Use when the user says "session budget", "should I compact", "should I clear", "is this session getting heavy", "token budget", invokes /session-budget, or when the budget hook requests a proactive assessment.
Install
/plugin marketplace add borkweb/skillsThe repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.
npx skills add borkweb/skills --skill session-budgetgit clone --depth 1 https://github.com/borkweb/skills.git /tmp/skills
mkdir -p ~/.claude/skills
cp -r /tmp/skills/skills/core/session-budget ~/.claude/skills/session-budgetPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Session Budget
Decide whether this session should be compacted or cleared, and if so, hand the
next session a clean briefing. Size alone is never the reason — a large session
mid-thought should be left alone. The real question is **composition**: how much of
what you are carrying is dead weight.
## 1. X-ray the context
Assess the context currently available to you; introspection and transcript bytes are approximate signals, not exact active-context measurements. Prefer runtime context metrics when exposed, and do not infer active size from an old or compacted transcript. Assess by introspection — do **not**
re-read the transcript. Report three things, briefly:
- **Reclaimable vs. load-bearing.** Roughly what fraction of current context is
stale — finished sub-tasks, abandoned exploration, superseded tool output, large
dumps no longer referenced — versus still live for the work ahead.
- **Seam.** Are you at a natural breakpoint, or mid-thought with interdependent
state a lossy handoff would damage?
- **Size signal.** One cheap absolute measurement to ground the estimate, if a
transcript path is available:
```bash
wc -c "${CLAUDE_TRANSCRIPT_PATH:-/dev/null}" 2>/dev/null
```
If unavailable, say so and rely on the composition estimate alone.
## 2. Give a verdict
Be honest, not agreeable. The user usually runs this *hoping* the answer is "yes,
compact" — don't reward that hope. **NOT YET is a common and correct answer**; giveDetails
- Skill name
session-budget- Repository
- borkweb/skills
- Path
skills/core/session-budget/SKILL.md- License
- MIT
- Repo updated
- 18 November 2026
FAQ
What does the Session Budget skill do?
Assess whether compacting or clearing the current session is worth it right now, and whether a reinit-ready handoff that survives /clear would carry real value. Not a threshold nag — it X-rays context composition (how much is stale vs. still load-bearing), which only the agent can see. Use when the user says "session budget", "should I compact", "should I clear", "is this session getting heavy", "token budget", invokes /session-budget, or when the budget hook requests a proactive assessment.
How do I install the Session Budget skill?
Claude Code plugin: /plugin marketplace add borkweb/skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Session Budget popular?
It lives in borkweb/skills, which has 9 GitHub stars and 0 forks.
Related skills
Prototype
9borkweb/skills
Build a throwaway prototype to flesh out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, explore design options, or says "prototype this", "let me play with it", "try a few designs".
Agents Md Lint
9borkweb/skills
Audit and trim AI agent instruction files (AGENTS.md, CLAUDE.md, CONVENTIONS.md, .cursorrules, etc.) by testing which facts an AI agent can discover from code alone. Use when asked to lint, audit, optimize, prune, or trim any agent instruction file in a repository. Removes redundant documentation that wastes context tokens. Supports a dry-run mode that reports findings without modifying files.
Auto Scope
9borkweb/skills
Scopes a coding task before implementation — names the handful of files actually relevant (read/edit these), the areas to leave alone, and any open scope questions, so work doesn't start by reading the whole repo. Fires proactively at the start of substantive, multi-file, or unfamiliar-area work; stays silent on single-file edits, one-line fixes, questions, and continued work in an area already scoped this session. Use when beginning a feature, bug fix, or refactor that could plausibly touch several files or an area not yet seen this session.
Agent Framework Azure AI Py
9huytran1120/vietnam-agentic-skills
Build persistent agents on Azure AI Foundry using the Microsoft Agent Framework Python SDK.
Desk
9borkweb/skills
Start, watch and answer bork `do` runs from a coordinating Claude session (in a herdr tab or over Remote Control); a Codex session inside its sandbox can only wait, show, ack and sometimes answer. Use to start a do run in a repo, wait for runs without polling, see which are done, in progress, blocked or need the user, and relay and answer their needs-me questions. Not for the backlog or Crew; matt:orchestrator owns those.
Agent Orchestrator
9huytran1120/vietnam-agentic-skills
Meta-skill que orquestra todos os agentes do ecossistema. Scan automatico de skills, match por capacidades, coordenacao de workflows multi-skill e registry management.
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.