← Back to Skills Marketplace
74
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install pm-skill
Description
强制指令：扮演产品经理（PM）与架构师角色。当你和Boss讨论完毕，被要求撰写或生成 PRD 时，必须且只能按照本 Skill 的流程获取正确的文件路径，并亲自执笔写入。
Usage Guidance
This skill is plausible for automating PRD creation, but it relies on and instructs the agent to run external scripts and modify files in your ~/.openclaw workspace and in another skill (leio-sdlc) without declaring those dependencies. Before installing or enabling it: 1) Verify the existence and contents of spawn_auditor.py and ~/.openclaw/skills/leio-sdlc/scripts/commit_state.py — review them for side effects. 2) Run the skill in an isolated or test workspace (not your production repo) to observe what files it touches. 3) Ensure templates and project paths referenced by init_prd.py actually exist where the skill expects them, or modify the skill to use a safe, declared config path. 4) If you allow autonomous invocation, restrict it until you confirm those external scripts are trustworthy. If you cannot inspect the external scripts, treat this skill as risky and avoid granting it write/exec permissions on real repositories or sensitive data.
Capability Analysis
Type: OpenClaw Skill
Name: pm-skill
Version: 1.0.0
The bundle contains a path traversal vulnerability in 'scripts/init_prd.py' because the '--project' argument is not sanitized before being used in 'os.path.join', potentially allowing the agent to create directories and write files outside the intended workspace. Furthermore, 'SKILL.md' contains highly prescriptive instructions (Prompt Injection surface) that mandate the execution of local scripts, including some located in external skill paths (e.g., 'leio-sdlc'), and 'preflight.sh' uses a risky pattern of executing any script matching 'scripts/test_*' without validation.
Capability Tags
Capability Assessment
Purpose & Capability
The skill claims to be a PM/PRD generator and contains code to create and edit PRDs, which is coherent. However, the runtime instructions require calling scripts located in ~/.openclaw/workspace/projects/... and explicitly require using another skill's commit script (~/.openclaw/skills/leio-sdlc/scripts/commit_state.py) and an auditor (spawn_auditor.py). Those external cross-skill calls are not declared in the metadata (no required config paths or env vars) and grant the skill implicit access to other parts of the user's workspace, which is disproportionate to a standalone PM prompt.
Instruction Scope
SKILL.md mandates using exec to run an absolute-path init_prd.py, then to read/edit/write the returned absolute file path. It also mandates immediately invoking spawn_auditor.py after writing and using a remote commit_state.py gateway to baseline the PRD. spawn_auditor.py and the referenced commit_state.py are not included in the package; instructing the agent to run unknown external scripts and to write files in other projects expands the skill's scope and could cause unintended side effects.
Install Mechanism
No install spec (instruction-only packaging) — low install risk. All code is bundled in the skill package (scripts, preflight, deploy), and there are no downloads from external URLs. Minor build/deploy scripts reference files (e.g., agent_driver.py) that aren't present in the manifest, which will cause runtime errors but not necessarily a security issue.
Credentials
The skill declares no required env vars or config paths, yet the runtime instructions assume access to the user's ~/.openclaw workspace, project directories, and another skill's scripts. This is a mismatch: the skill needs filesystem and cross-skill access but does not declare or limit it. That lack of explicit declaration increases the risk of accidental access or privilege overreach.
Persistence & Privilege
The skill is not always-on and allows autonomous invocation (platform default). It instructs the agent to save PRDs and to call a central commit_state.py to baseline files (modifying project state). While not flagged as 'always', calling cross-skill commit utilities grants write-side effects across the workspace and should be reviewed before permitting autonomous runs.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install pm-skill - After installation, invoke the skill by name or use
/pm-skill - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial v1.0 release for PRD authoring
Metadata
Frequently Asked Questions
What is PM Skill?
强制指令：扮演产品经理（PM）与架构师角色。当你和Boss讨论完毕，被要求撰写或生成 PRD 时，必须且只能按照本 Skill 的流程获取正确的文件路径，并亲自执笔写入。 It is an AI Agent Skill for Claude Code / OpenClaw, with 74 downloads so far.
How do I install PM Skill?
Run "/install pm-skill" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is PM Skill free?
Yes, PM Skill is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does PM Skill support?
PM Skill is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created PM Skill?
It is built and maintained by leio9511 (@leio9511); the current version is v1.0.0.
More Skills