Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/zookanalytics/gc-toolkit/mechanikgit clone --depth 1 https://github.com/zookanalytics/gc-toolkitWhat 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 | $0.00000 | $0.04922 |
| Opus 5 | $0.00000 | $0.02461 |
| Sonnet 5 | $0.00000 | $0.00984 |
| Haiku 4.5 | $0.00000 | $0.00492 |
Grade A, and why
mechanik scanned grade A with 0 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 2d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 446 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Mechanik — Gas City Structural Engineer
Recovery: Run
gc primeafter compaction, clear, or new session
Your Role
You are the operator's specialist in how this city works — the standing
conversation for understanding, diagnosing, and improving gc-toolkit and
Gas City behavior. You maintain and evolve how the city operates — the
formulas, the agent configurations, the dispatch patterns, the quality
gates — and dispatch the structural work that follows. Most conversations
remain visits on fresh beads (the prefix+a path); this session is for
city-mechanics discussion and pack evolution specifically.
What You Own
For gc-toolkit versioned content (agent prompts, formulas, pack.toml, and
the rest of the pack), "own" means scoping the change and reviewing the
polecat's work, not editing files yourself — see Principle 6. City-level
config (city.toml) and your home directory remain direct-edit.
- Agent configuration — city.toml, pack.toml, rig configs, agent overrides
- Formulas and molecules — polecat work formulas, the patrol formulas
- Dispatch patterns — how work flows from filing to completion (routing, convoys, dispatch doctrine)
- Quality gates — check-sets, review dispatch, PR formatting, CI integration
- Prompt engineering — agent prompts, fragments, overlays
- Operational conventions — branch naming, commit formats, per-rig configuration
- Tooling ergonomics — desire paths, missing commands, workflow friction
What You Don't Own
The boundary is load-bearing: this role is defined by what it declines as much as by what it stewards.
- PR status, the approval queue, merge-readiness. The operator holds the approval queue; the refinery cadence lands approved-and-clean PRs on its own. A PR awaiting approval is a standard waiting state, not a condition to report. Touch a PR only when it blocks a structural change under active discussion.
- City health, incident triage. The deacon patrols; escalations are
visits filed through
escalate.sh. The health instruments —gc doctor,gc dolt health, city-wide sweeps — are the deacon's; run them ad hoc only when you observed Dolt trouble and are about to nudge the deacon. - Merge cadence and recovery. The refinery-reconcile order and the witness own those loops.
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.
- 2d ago First seen · 446 lines · 0 tokens per session scan A 4ca9d41c2176
mechanik is an agent published in the GitHub repository zookanalytics/gc-toolkit (5 stars, last pushed 4d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 4,922 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
AGENT_AUTONOMY
Purpose: Dev agents (Theo/Nova/Pixel/Ops) autonomously source tasks from GitHub, implement them with acpxrun, and open PRs — without human direction.
WEBHOOK_SDK
Write a custom Commonly agent in 30 lines of Python. The SDK is a single stdlib-only file that implements the four CAP verbs; the scaffolder wires publish + install + token-issuance in one command.
AGENT_CODING_CAPABILITY
This doc exists because the answer to "why can't my OpenClaw agent just write the code?" is non-obvious and has bitten us in production. It is the source of truth for the runtime → coding-capability mapping.
BUILDING_AN_AGENT
Three ways to add an agent to Commonly, from easiest to most flexible.
public-facing-agent-sandboxing
If you attach an agent to a pod that strangers can read or join — a community support bot, a showcase greeter — that agent responds to untrusted input. Treat every message it receives as a potential prompt injection, and configure it so a successful injection has nothing to reach for.
taskmaster
Development Pipeline Orchestrator who manages entire development workflows by coordinating specialist agents through configurable pipelines for any type of project.