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/yunsii/wezdecknpx agentmods add skills/yunsii/wezdeck/dev-taskWrote 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/yunsii/wezdeck/dev-task)<a href="https://agentmods.dev/skills/yunsii/wezdeck/dev-task"><img src="https://agentmods.dev/badge/skills/yunsii/wezdeck/dev-task/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/yunsii/wezdeck/dev-task"><img src="https://agentmods.dev/badge/skills/yunsii/wezdeck/dev-task.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 | $0.00069 | $0.02329 |
| Opus 5.5 | $0.00028 | $0.00932 |
| Sonnet 5.5 | $0.00014 | $0.00466 |
| Haiku 4.5 | $0.00007 | $0.00233 |
Grade A, and why
dev-task 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 20d 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 — 191 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dev task (allowlisted repos)
When
Write work only under allowlist (see AGENTS.md). Pure Q&A: skip. Other repos: refuse.
| Logical | Roots |
|---|---|
| 团队仓 | $HOME/work/team-repo, $HOME/work/.worktrees/team-repo |
| wezdeck | $HOME/github/wezterm-config (primary), optional .worktrees |
cwd policy (L0-12/13)
Core: personal projects → default mainline. wezdeck is the local instance.
| Repo | Default | Worktree when |
|---|---|---|
| wezdeck (personal) | primary master |
parallel, long experiment, isolated agent cwd, user asks |
| 团队仓 (team-ish) | claw-* under dirname(primary)/.worktrees/<repo>/ |
default isolation unless user overrides |
Path formula (when used): dirname(realpath(primary))/.worktrees/<basename(primary)>/<slug>/.
Architecture: openclaw/docs/agent-architecture.md.
Checklist
Same steps as AGENTS.md Write-task checklist (wezdeck may skip assess/create).
Scripts: dev-task-ledger.sh, claw-worktree.sh, claw-run.sh.
Assess → create (only when isolation/parallel needed)
./openclaw/scripts/claw-worktree.sh assess \
--title "<subject>" --domain "<area>" --scope "<hint>" [--days N] \
--cwd "$HOME/github/wezterm-config"
./openclaw/scripts/claw-worktree.sh create \
--title "…" --lifecycle task|dev|hotfix --domain "…" \
--cwd "$HOME/github/wezterm-config"
- Prefer-reuse same domain; never human
dev-*/task-*as write targets. - Reclaim never automatic after close (only if a worktree was used).
Rails & modes (user-facing)
| 轨 | 方式 | 旧 | Who codes | Main does |
|---|---|---|---|---|
| 人工 | H1 人直接 | A | User IDE | Ledger/验收 only |
| 人工 | H2 原生 Agent | A | Host grok/claude/codex | Assist only |
| Claw | C1 Main 自写 | B | Main (Main-Grok) | Implement + verify |
| Claw | C2 Handoff | C | Host CLI after handoff | Stop coding that cwd |
| Claw | C3 ACP 后端 | E | ACP → claude | codex | Spawn/close; single writer |
| Claw | Ticket-headless | — | delegate host headless |
跑同一 skill cross-repo-delegate；票在 ~/.agent/tickets/ |
| — | D | D | — | Forbidden |
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.
- 20d ago Changed d85900e5453c
- 1mo ago Changed · +5 lines 1ffac98c7340
- 1mo ago First seen · 186 lines · 69 tokens per session scan A f7a65346e0b8
dev-task is a skill published in the GitHub repository yunsii/wezdeck (5 stars, last pushed yesterday), licensed MIT. It adds 69 tokens to every session and 2,329 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
author-contributions
Identify all files a specific author contributed to on a branch vs its upstream, tracing code through renames. Use when asked who edited what, what code an author contributed, or to audit authorship before a merge. This skill should be run as a subagent — it performs many git operations and returns a concise table.
commit
Commit staged or unstaged changes with an AI-generated commit message that matches the repository's existing commit style. Use when the user asks to 'commit', 'commit changes', 'create a commit', 'save my work', or 'check in code'.
release-announcement
Write a release announcement — changelog, blog post, in-app note, or social post — that leads with user impact, names the audience, and includes upgrade/migration steps without filler.
github-pr-workflow
Prepare a GitHub pull request from a feature branch — branch hygiene, commit shape, title/body, verification notes, screenshots for UI work, and replies to review comments.
contributor-onboarding
Help a new contributor get productive on this checkout - inspect sync state against main, build, run the repository's exact verification gate, and produce a local what's-new digest. Never fetches, pulls, or modifies a dirty tree on its own. Explicit-only.
codew-release-qa-sweep
Use before claiming Codewhale release work is done: run the full gate sweep and list the manual QA targets.