BF wechat-use
macOS WeChat CLI + local HTTP bridge + Wechaty Puppet gRPC gateway — send messages, query sessions / contacts / chat history / images / favorites, and expose stable HTTP / gRPC surfaces for agent integration. Use when the user asks to 'send a WeChat message', '发微信', query WeChat contacts/groups/messages, look up who said what in a chat, fetch images from history, export chat history, wire WeChat into Hermes / n8n / Dify / LangChain, or run any wechaty bot on a real macOS WeChat account. Requires WeChat 4.1.8 / 4.1.9 on macOS (Apple Silicon) and a `wechatuse_` activation code. One-time `wechat-use init` extracts the DB key; no sudo, no re-signing WeChat.app. Optional remote bridge — `wechat-use tunnel setup --hostname <yours>` exposes the local REST API via Cloudflare Tunnel for remote services to call.
macOS WeChat CLI + local HTTP bridge + Wechaty Puppet gRPC gateway — send messages, query sessions / contacts / chat history / images / favorites, and expose…
As a process F 62/100 · Will not run — References files that are not bundled: LICENSE
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- The text references files that are not there: add them or drop the references.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 6
✓ No critical or high findings
Medium and low: 6
-
medium Dangerous commands
cmd-pipe-to-shell-known-hostSKILL.md:230Pipe-to-shell installer from a well-known host (still executes remote code)curl -fsSL https://raw.githubusercontent.com/leeguooooo/wechat-use/main/install.sh | bash
-
medium Dangerous commands
cmd-pipe-to-shell-known-hostSKILL.md:722Pipe-to-shell installer from a well-known host (still executes remote code)curl -fsSL https://raw.githubusercontent.com/leeguooooo/wechat-use/main/install.sh | bash
-
low Dangerous commands
cmd-shell-rcSKILL.md:234Writes to a shell startup file (detector / deny-list definition)zsh) grep -q '.local/bin' ~/.zshrc 2>/dev/null || echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.zshrc ;;
detector -
low Dangerous commands
cmd-shell-rcSKILL.md:235Writes to a shell startup file (detector / deny-list definition)bash) grep -q '.local/bin' ~/.bashrc 2>/dev/null || echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc ;;
detector -
low Dangerous commands
cmd-pipe-to-shell-known-hostSKILL.md:288Pipe-to-shell installer from a well-known host (still executes remote code) (documentation table row)| `daemon_accessibility FAIL` / `ax_trusted=false` after a recent install/upgrade | Multiple `ai.wechat.*` LaunchAgents on the machine — orchestrate (KeepAlive) re-spawned wechatd with a stale launchd
table -
low Dangerous commands
cmd-persistenceSKILL.md:295Persistence mechanism (cron / launchd / scheduled task / autorun registry) (security demo / example; quoted — discussed, not commanded)**LaunchAgent / TCC handbook for agents**: if you ever need to manually reset responsibility chains (e.g. user's machine has a customized LaunchAgent layout), use `launchctl bootout` + `launchctl boot
demoquoted
Files scanned: 3. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 10991 tokens (recommended < 5000); move details to references/ - warning
missing-refreference to a missing file: LICENSE - note
edit-residuethe text marks something as outdated (lines 98, 562): check that old rules are not kept next to new ones — the full check reads the text for contradictions
Process rating: all ten parameters 62/100
- 0Tools and files. 1 referenced file(s) missing: LICENSE
- 0Inputs and preconditions. Does not say what the process needs to start
- 40Execution cost. Instruction body is 10991 tokens: crowds the task out of the window
- 60Result and completion. Output format stated, no completion criterion
- 100Steps. 57 steps
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 3 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- low 16 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (32 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 813: 120–800 characters recommended
- -2localhost URLs: will not work for another user
- -230 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +4Structure: 27 headings
- +3Step-by-step instructions: 57 items
- +3Output format is stated explicitly
- +4Has examples (25 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 64.