Borrowing it
Nothing to install: this file belongs to lbruton/claude-context. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/lbruton/claude-context/master/.agents/skills/start/SKILL.mdgit clone --depth 1 https://github.com/lbruton/claude-contextWrote 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/lbruton/claude-context/start)<a href="https://agentmods.dev/skills/lbruton/claude-context/start"><img src="https://agentmods.dev/badge/skills/lbruton/claude-context/start/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/lbruton/claude-context/start"><img src="https://agentmods.dev/badge/skills/lbruton/claude-context/start.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00102 | $0.03734 |
| Opus 5.5 | $0.00041 | $0.01494 |
| Sonnet 5.5 | $0.00020 | $0.00747 |
| Haiku 4.5 | $0.00010 | $0.00373 |
Grade A, and why
start 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 5d 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 — 335 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Start
Lightweight session reorientation. Use this when you're continuing recent work and just need to know where you left off — not when you need a full environment boot.
Under 15 seconds. No code indexing, no agent dispatch. One Bash sweep for active work, one Plane list, and two fast memory calls (sessionflow + mem0).
Harness-neutral. One master in
Devops/skill-library/skills/start/, deployed to.claude/skills/(Claude Code) and.agents/skills/(Codex). Edit the master. MCP tool names below name capabilities; resolve the live identifier from your tool catalog (Claude Code:ToolSearch), since some prefixes are connection-specific.
Step 0 — Verify project identity (MUST run first, blocking)
Read .claude/project.json (relative to cwd) with the file-read tool, and run:
pwd && git rev-parse --show-toplevel
If .claude/project.json is missing OR pwd differs from git rev-parse --show-toplevel
(e.g., launched from a subdir whose parent is a different git repo — common when a project
like Devops/ sits inside an outer workspace repo), STOP the flow and warn prominently.
Call the verified root ROOT; every later command anchors to it.
Step 1 — Git state snapshot
If the harness injected a gitStatus block at session start (Claude Code does), read the
current branch, uncommitted changes, and last 5 commits from it — it is authoritative for
the current checkout only. Otherwise run:
cd "$ROOT" && git status --short && git branch --show-current && git log --oneline -10
Only reach for GitHub (mcp__github__list_commits) if the user asks to verify or you
suspect stale context after a force-push / external merge.
Step 1b — Active work sweep (MUST run, every time)
The injected git block cannot see other worktrees, stashes, or work another harness or
machine pushed. Codex, another Claude account, or a cloud session may have work in flight
that never reached mem0 (no /wrap yet) and that the Step 3 keywords would not match.
Run this single Bash call:
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.
- 5d ago Changed · +4 lines 0c14ada08008
- 9d ago Changed e50e722587e3
- 11d ago First seen · 331 lines · 102 tokens per session scan A 61fedbf8b0aa
start is a skill published in the GitHub repository lbruton/claude-context (2 stars, last pushed 5d ago), licensed MIT. It adds 102 tokens to every session and 3,734 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-10-17.
Other skills, from other repositories
conventional-commit
Rules for writing Conventional Commits messages in this project. MUST be invoked every time a commit is created.
python-release
Handle Python SDK release, build, bump, packaging metadata, PyPI client pin, uv.lock, nox/build workflow, and publish verification changes. Use for Python release process work or dependency pin bumps; do not use for ordinary Python feature implementation.
release-notes
Generate and publish concise, evidence-based notes in the body of the latest existing GitHub Release. Use only when the user explicitly invokes $release-notes or explicitly asks to update the latest existing GitHub Release body. Do not invoke for general release planning, changelog, tag, or version tasks.
ratchet-files
Ratchet source file size so every file stays small enough for an agent to read whole: find every file over a token budget, split oversized files by concept, and set up the pre-commit ratchet that stops them from growing. Use when a commit is blocked because a file is over the size budget, when asked to find or split…
git-guardrails-claude-code
A Claude Code setup that blocks selected destructive Git commands before they run. Git is a tool for tracking code changes; the blocked commands include pushing, deleting branches, discarding changes, and hard resets.
ai-visibility-writing
Audit, question, suggest, or fact-preservingly revise a press release, blog post, contributed article, or expert explainer so AI answer systems can more easily retrieve, understand, quote, and cite its useful information. Use when someone asks for AI visibility, AI search, answer-engine, AEO, GEO, AI Overview, or…