av/harbor is a command-line tool and companion app that uses Docker Compose to start a connected local stack of language-model backends, user interfaces, and supporting AI services. People use it to run services such as Ollama, llama.cpp, vLLM, Open WebUI, search, voice, and image-generation tools without configuring their connections manually, while the catalogue provides agent workflows for operating Harbor.
Borrowing it
Nothing to install: this file belongs to av/harbor. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/av/harbor/main/.agents/skills/facts/SKILL.mdgit clone --depth 1 https://github.com/av/harborWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/av/harbor/facts)<a href="https://agentmods.dev/skills/av/harbor/facts"><img src="https://agentmods.dev/badge/skills/av/harbor/facts/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/av/harbor/facts"><img src="https://agentmods.dev/badge/skills/av/harbor/facts.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 4 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Tool Misuse · line 20 Tool calls are chained to bypass individual safety checks or escalate capabilities beyond what any single tool call would allow.Fix: Limit tool chaining depth and validate the output of each tool before passing it to the next. Require explicit user approval for multi-step chains.
- medium MCP Rug Pull · line 31 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium Excessive Agency · line 74 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- low Supply Chain · line 20 Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.Fix: Avoid downloading and executing remote scripts. Use trusted packages from PyPI/npm. If remote fetch is required, verify checksums and use HTTPS.
What 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 5.1 | $0.00049 | $0.02274 |
| Opus 5.5 | $0.00020 | $0.00910 |
| Sonnet 5.5 | $0.00010 | $0.00455 |
| Haiku 4.5 | $0.00005 | $0.00227 |
Grade C, and why
facts 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 2mo ago.
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.
curl -fsSL https://raw.githubusercontent.com/av/facts/main/install.sh | sh Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsSL https://raw.githubusercontent.com/av/facts/main/install.sh | sh How it starts
The opening of the file, as written. The whole thing — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.
facts
A CLI for fact-driven development. You use it to specify what must be true about a project, then validate that reality matches.
Project: https://github.com/av/facts
Installing
If facts is not installed, install it with one of:
curl -fsSL https://raw.githubusercontent.com/av/facts/main/install.sh | sh
npm install -g @avcodes/facts
pipx install facts-cli
Verify with facts --version. The command is always facts — never npx facts.
Core idea
A .facts file is a flat list of atomic truth statements about a project. Each fact can optionally have a shell command that verifies it. The fact sheet serves as both specification (what should be true) and documentation (what is true) — the difference is just which direction you're working from.
- the API returns JSON
- label: project builds
command: cargo build
- label: tests pass
command: cargo test
tags: [ci, core]
That's the entire format. Plain strings for simple facts, mappings when you need a command, tags, or explicit ID. Allowed mapping keys: id, label, command, tags — nothing else.
Essential commands
Recommended short aliases (all extra arguments are passed through):
ll=list --light— the most common "skim" view (markdown headings + dim IDs)ls=listrm=removeat <id> <tag>=edit <id> --add-tag <tag>(supports multiple IDs and--label/--new-idetc. after the tag)rt <id> <tag>=edit <id> --remove-tag <tag>
These are the highest-ROI shortcuts for daily and agent use.
See everything:
facts ll
facts ll --tags "not implemented"
facts ll --has-command
facts list --light # or the alias: facts ll
Validate:
facts check
facts check --tags "ci"
check is your primary feedback loop. It lints the files first (aborting on structural errors), then runs every command-fact and reports pass/fail/manual. Run it often. Exit 0 means all command-facts pass; manual facts don't affect the exit code.
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.
- 2mo ago First seen · 203 lines · 49 tokens per session scan C 2e4fcfbe392b
facts is a skill published in the GitHub repository av/harbor (3,263 stars, last pushed 6d ago), licensed Apache-2.0. It adds 49 tokens to every session and 2,274 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-08-30.
facts compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| factsav/harbor | Skill | 3.3k | 6d ago | Apache-2.0 | C | 49 | Claude Code, Codex |
| recover-contextpchalasani/claude-code-tools | Skill | 2.0k | yesterday | MIT | A | 36 | Claude Code |
| lc-curate-contextcyberchitta/llm-context.py | Skill | 308 | 18d ago | Apache-2.0 | A | 90 | Claude Code, Codex |
| obsidianlvndry/jazz | Skill | 55 | yesterday | MIT | A | 71 | Claude Code, Codex |
| difyznlgis/opengis-skills | Skill | 76 | 10d ago | MIT | A | 115 | Claude Code, Codex |
| session-searchpchalasani/claude-code-tools | Skill | 2.0k | yesterday | MIT | C | 33 | Claude Code |
| knowledge-session-state-and-context-active-turn-interactions-and-recoveryechoVic/blade-code | Skill | 181 | 24d ago | MIT | A | 161 | Claude Code, Codex |