Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/beevibe-ai/beevibenpx agentmods add skills/beevibe-ai/beevibe/beevibe-pre-task-setupWrote 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/beevibe-ai/beevibe/beevibe-pre-task-setup)<a href="https://agentmods.dev/skills/beevibe-ai/beevibe/beevibe-pre-task-setup"><img src="https://agentmods.dev/badge/skills/beevibe-ai/beevibe/beevibe-pre-task-setup.svg" alt="Measured on agentmods" 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 | $0.00198 | $0.01250 |
| Opus 5 | $0.00099 | $0.00625 |
| Sonnet 5 | $0.00040 | $0.00250 |
| Haiku 4.5 | $0.00020 | $0.00125 |
Grade A, and why
beevibe-pre-task-setup 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 6d 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pre-Task Setup
When this fires
Every session that starts with a <task id="..."/> intent block. The session lifecycle (per the beevibe umbrella) requires you to set up your worktree before doing any work.
The protocol
1. Find the repo URL
Check for repo_url on the task. Fetch the task with get_task(task_id) if you don't already have it. The repo_url field tells you where to clone.
If task.repo_url is null:
- Parse the task description for git context (URLs, branch hints, "the foo repo")
- If still ambiguous, call
report_blocker(task_id, "Cannot determine repo for this task")— the parent agent or operator will clarify
2. Check for existing clone
Your cwd is the workspace root (~/.beevibe/workspaces/<your_agent_id>/). The base clone lives at <repo_name>/:
ls <repo_name>/.git/HEAD 2>/dev/null && echo "EXISTS" || echo "MISSING"
3. Clone or pull
- If missing:
git clone <task.repo_url> - If present: refresh the base:
cd <repo_name> git fetch --prune origin git checkout <default_branch> git pull --ff-only
4. Prune landed worktrees from earlier tasks
Before adding a new worktree, reclaim sibling <repo_name>-<task_id_short> dirs from earlier tasks whose work has already landed on origin/<default_branch>. git cherry compares by patch-id, so this catches both merge-commit and squash-merge cases:
# Run from inside <repo_name>/ (the base clone).
for wt in $(git worktree list --porcelain | awk '/^worktree/ {print $2}'); do
branch=$(git -C "$wt" symbolic-ref --short HEAD 2>/dev/null || echo "")
case "$branch" in
agent/*)
# 0 unmerged commits → all work already on the default branch → safe to remove.
if [ "$(git cherry origin/<default_branch> "$branch" 2>/dev/null | grep -c '^+')" = "0" ]; then
git worktree remove "$wt" 2>/dev/null && git branch -D "$branch" 2>/dev/null
fi
;;
esac
done
Worktrees with uncommitted state or unmerged commits stay — those may be pending revision or post-escalation. No --force here on purpose: if git worktree remove refuses, the worktree was still doing something useful.
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.
- 6d ago First seen · 105 lines · 198 tokens per session scan A 6ac2e9a82f48
beevibe-pre-task-setup is a skill published in the GitHub repository beevibe-ai/beevibe (345 stars, last pushed yesterday), licensed Apache-2.0. It adds 198 tokens to every session and 1,250 once invoked, about $0.0010 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
kungfu-agent-onboarding
Discover the exact Kungfu Project, WorkConsole, WorkRef, Skill catalog, and Core Work state admitted to this Amp process.
kungfu-agent-onboarding
Use when a user asks to understand, start, inspect, extend, or safely operate installed Kungfu; verify the installed pack, select one intent route, personalize the explanation, and propose one smallest safe next action.
kungfu-agent-onboarding
Use when a user asks to understand, start, inspect, extend, or safely operate installed Kungfu; verify the installed pack, select one intent route, personalize the explanation, and propose one smallest safe next action.
pr-body
分析当前分支的全部变更,自动生成或更新 PR 标题和正文。聚焦内容质量:解释 why、归纳 what、标注测试方式,遵循 anolisa 项目 PR 模板规范。适用于新建 PR 前生成描述,或已有 PR 需要更新描述。.
github-ops
GitHub CLI for issues, PRs, code search, CI logs, releases, and API queries. Requires gh CLI and auth.
release-cut
Cut a new pi-agent-dashboard release: promote ## [Unreleased] in CHANGELOG.md, bump every workspace package.json per SemVer, commit, tag v , and push — triggering the Release workflow that publishes every non-private workspace, builds the Electron artifacts, and creates a GitHub Release. Use on "cut a release"…