BB openclaw_projects
Builds a Project — a coordinated multi-agent team setup — inside OpenClaw, for any kind of team: software development, marketing, real estate, content, sales, operations, research, customer success, or anything else where multiple agents need to work together toward shared goals. Use this skill whenever someone says "set up a project", "create a project", "add a project to my team", "build a team", "make my agents work together", "configure agent coordination", "set up agent collaboration", "I want a team of agents", "how do I run multiple agents on one project", "wire up Asana for my agents", "wire up ClickUp for my agents", "add a new project", or anything similar — even if they don't explicitly say "project" or "team". Also trigger when someone asks how multiple agents should coordinate, share work, escalate, or hand off tasks. This skill creates the entire project folder structure (PROJECT.md rulebook, project.json config, queue files for inter-agent messaging, project-lock.json phase tracking, decision/issue/runbook documents, shared workspace) and updates OpenClaw config to wire agent-to-agent communication. It works through a structured interview: team identity, work structure, then a comprehensive AI-rewritten team plan for user review and fine-tuning before building. Supports any task manager backend (Asana, ClickUp) via the user's separately installed dependency skill. Multiple projects can coexist; agents can participate in multiple projects. Requires the openclaw-administrator skill (EncryptShawn) to be loaded. Recommends openclaw-recovery-manager (EncryptShawn) for safety. This skill does not make task-manager API calls itself — those are delegated to the user's installed Asana or ClickUp dependency skill. This skill does not read or store any credentials or secret values.
As a process B 71/100 · Nearly there — weak spots: consistency
How to improve
- Shorten the description to 1024 characters.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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 · 0
✓ No critical or high findings
Files scanned: 8. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1817 chars, limit 1024 - warning
name-formatname should be kebab-case (lowercase letters, digits, hyphens) - warning
body-longSKILL.md body ≈ 7435 tokens (recommended < 5000); move details to references/ - note
description-budgetdescription takes 1817 of the ~15000-char shared budget for all skills
Process rating: all ten parameters 71/100
- 40Consistency. Frontmatter name (openclaw_projects) differs from the folder (openclaw-projects)
- 60Tools and files. Uses tools (web, git) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70When it triggers. States when to use, but not when not to
- 70Inputs and preconditions. Inputs and preconditions are listed
- 70Execution cost. Instruction body is 7435 tokens
- 85Steps. 93 steps, 1 vague phrases
- 100Failures and branches. 19 branches, has a failure section
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 20 top-level sections: this looks like several domains in one skill
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
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1816: 120–800 characters recommended
- +1No license
- +2Single-language instructions
- +5Description quotes 13 example trigger phrases
- +4Structure: 32 headings
- +3Step-by-step instructions: 93 items
- +3Output format is stated explicitly
- +4Has examples (20 code blocks)
- +4Reference files are cited in the instructions (5 of 5)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 50.