Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/mendcode/mendcode/mendcode-release-flownpx skills add MendCode/MendCode --skill mendcode-release-flowgit clone --depth 1 https://github.com/MendCode/MendCodeWhat 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 | $0.00085 | $0.01197 |
| Opus 5 | $0.00043 | $0.00598 |
| Sonnet 5 | $0.00017 | $0.00239 |
| Haiku 4.5 | $0.00009 | $0.00120 |
Grade C, and why
mendcode-release-flow scanned grade C with 2 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 yesterday.
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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
- For installer: test `curl -fsSL https://raw.githubusercontent.com/MendCode/MendCode/main/src/mendcode/install | bash -s -- --no-modify-path` in a temporary `HOME`. Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- For installer: test `curl -fsSL https://raw.githubusercontent.com/MendCode/MendCode/main/src/mendcode/install | bash -s -- --no-modify-path` in a temporary `HOME`. How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MendCode Release Flow
Follow this workflow for MendCode changes that may ship.
Non-negotiables
- For maintainer/agent work inside the main MendCode repository, use the local
devbranch by default. Do not create a separate branch unless the user explicitly asks for one, the current worktree has unrelated changes in the same files, or a GitHub contribution flow requires it. - Preserve unrelated local changes. If the active checkout is dirty, inspect the touched files first. Work in-place on
devonly when the requested files are clean or the existing edits are clearly part of the same user-approved task. - If unrelated dirty files conflict with the task, stop and explain the blocker before creating a worktree or branch.
- Before editing release/version/package files, ask or verify whether another agent already bumped the version locally or remotely.
- Do not publish a release, merge to
main, or overwrite a user worktree without explicit user intent. - Never leave open public PRs/issues as noise. Close, merge, or explain exactly why they must remain open.
- Never hide security/release failures. If a supply-chain, secret, CodeQL, release, or installer check fails, keep the issue open until fixed or document the exact reason it is accepted.
Branch Policy
- Internal MendCode maintainers and local agents work on
devby default. - External contributors should use a fork or feature branch and open a PR targeting
dev, notmain. mainis the public release branch. Only promotedevtomainafter CI passes, the user-visible change has been tested, version/changelog state is correct, and the user intends to ship.- For internal promotion, prefer a direct
dev->mainPR. Do not createcodex/*promotion branches unless GitHub cannot represent the intended merge directly or a conflict must be resolved outside the user's local checkout. - If an emergency promotion branch/worktree is unavoidable, delete the remote branch, local branch, and temporary worktree immediately after merge or close.
- Local release or promotion workspaces under
$TMPDIR/mendcodemust be run throughsrc/mendcode/script/releaseor its sourced cleanup helper so the current workspace'ssrc/mendcode/node_modulesis removed on success or failure. Do not run broad age-based temp cleaners or delete unowned/active temp workspaces. - Do not merge random branches directly into
main. Bring useful branches back through PRs todev, then delete stale branches after merge. - Before starting, check whether another PR/agent already contains the same work. Prefer continuing the existing local
devwork over creating a duplicate branch.
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.
- yesterday First seen · 67 lines · 85 tokens per session scan C 0f91f1f58aba
mendcode-release-flow is a skill published in the GitHub repository MendCode/MendCode (5 stars, last pushed 3d ago), licensed MIT. It adds 85 tokens to every session and 1,197 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
agent-harness-fault-injection
Use when an agent workflow needs deterministic recovery evidence for sandbox, MCP/tool, worker, checkpoint, memory, or orchestration failures.
verify-pr
This skill should be used to run a sandboxed deep verification of a qwen-code PR — "/verify-pr ", "深度验证这个 PR", A/B load-bearing proof against the base build, mock-free harnesses with wire oracles, and targeted gates — producing tmp/pr -verify- /report.md plus a machine-readable verdict. Designed for the token-free CI…
stuck
Diagnose frozen, stuck, or slow Qwen Code sessions on this machine. Scans for problematic processes, high CPU/memory usage, hung subprocesses, and debug logs. Use /stuck or /stuck to focus on a specific process.
e2e-testing
Guide for running end-to-end tests of the Qwen Code CLI, including headless mode, MCP server testing, and API traffic inspection. Use this skill whenever you need to verify CLI behavior with real model calls, reproduce user-reported bugs end-to-end, test MCP tool integrations, or inspect raw API request/response…
structured-debugging
Hypothesis-driven debugging methodology for hard bugs. Use this skill whenever you're investigating non-trivial bugs, unexpected behavior, flaky tests, or tracing issues through complex systems. Activate proactively when debugging requires more than a quick glance — especially when the first attempt at a fix didn't…
desktop-brand-builder
Generate a branded Qwen Code desktop package from the Tauri desktop shell using a minimal brandId and logo. Use when the user wants a custom, white-label, or rebranded desktop client, installer, DMG/EXE/AppImage/deb, or one-click brand build on top of packages/desktop-shell.