tachi — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited tachi (Agent Skill) and scored it 82/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 2 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 2 flagged
The text {match} is the classic direct prompt-injection phrasing. Placed in a skill body that the agent reads as trusted instructions, it tries to make the agent abandon its prior rules and follow whatever comes next — a full system-prompt override.
ignore/disregard/forget … previous instructions sentence.The text {match} is the classic direct prompt-injection phrasing. Placed in a skill body that the agent reads as trusted instructions, it tries to make the agent abandon its prior rules and follow whatever comes next — a full system-prompt override.
ignore/disregard/forget … previous instructions sentence.Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Show available TachiBot skills, tools, and API key status.
/tachi
/tachi statsWhen user invokes /tachi:
Present this overview:
## TachiBot - Multi-Model AI Orchestration
### Skills (slash commands)
| Skill | What it does | Example |
|-------|-------------|---------|
| `/judge` | Multi-model council - parallel analysis + synthesis | `/judge how to implement auth` |
| `/think` | Sequential reasoning chain with any model | `/think grok,gemini design a cache` |
| `/focus` | Mode-based reasoning (debate, research, analyze) | `/focus architecture-debate SQL vs NoSQL` |
| `/breakdown` | Strategic overview: first principles + feasibility | `/breakdown add real-time collaboration` |
| `/decompose` | Split into pieces, deep-dive each one | `/decompose implement auth system` |
| `/prompt` | Pick the right thinking technique for your problem | `/prompt why do users churn` |
| `/algo` | Algorithm analysis with 3 specialized models | `/algo optimize LRU cache O(1)` |
| `/tachi` | This help screen | `/tachi` |
### Quick Start
1. Pick a skill based on your task:
- **Need an answer?** -> `/judge`
- **Need to reason through something?** -> `/think`
- **Need to debate approaches?** -> `/focus architecture-debate`
- **Need a strategic overview?** -> `/breakdown`
- **Need to understand each piece deeply?** -> `/decompose`
- **Need a thinking technique?** -> `/prompt` (recommends one for you)
- **Algorithm or optimization?** -> `/algo`
2. Or use tools directly: `grok_search`, `gemini_brainstorm`, `qwen_coder`, etc.Call the usage_stats tool to show what's configured:
mcp__tachibot-mcp__usage_stats({ action: "view", scope: "current", format: "table" })Then summarize which providers are active based on whether their tools appear in available MCP tools.
/tachi stats:Call usage stats with full scope:
mcp__tachibot-mcp__usage_stats({ action: "view", scope: "all", format: "table" })Present the usage data with insights about most-used tools and cost efficiency.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.