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 skills add wiber/thetacog-mcp --skill steergit clone --depth 1 https://github.com/wiber/thetacog-mcpWrote 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/wiber/thetacog-mcp/steer)<a href="https://agentmods.dev/skills/wiber/thetacog-mcp/steer"><img src="https://agentmods.dev/badge/skills/wiber/thetacog-mcp/steer/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/wiber/thetacog-mcp/steer"><img src="https://agentmods.dev/badge/skills/wiber/thetacog-mcp/steer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00126 | $0.01832 |
| Opus 5.5 | $0.00050 | $0.00733 |
| Sonnet 5.5 | $0.00025 | $0.00366 |
| Haiku 4.5 | $0.00013 | $0.00183 |
Grade A, and why
steer 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 yesterday.
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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/steer — away from home, through the installed doors only
This is the STRANGER-SAFE /steer. It ships inside the ThetaCog extension
(packages/thetacog-mcp-vscode/skills/steer/SKILL.md) and is meant to run in ANY repo
the extension is installed into — never only inside the extension's own development
repo. Everything it runs is a door: npx thetacog-mcp <door> [args], the published
package's own dispatch table, never a command that only exists on one machine. If a
door is not listed below, it does not exist here yet — say so (not run — no <name> door) rather than guessing a command that is not present in this workspace.
THE CHAIR NEVER BLOCKS THE THREAD (HARD RULE, operator 2026-11-04)
Verbatim: "use subagents with the /steer - you bloced this thread (make /steer never do that -". The chair's own calls stay bounded — under ~30 s. Any test suite, compile, walk, render, or worker that can run longer goes to run_in_background or a subagent (one Agent call per independent job, all dispatched in a single message when the jobs are independent). While they run, the chair keeps answering the operator — it never sits silent waiting on its own foreground call. A foreground call that hit a timeout anyway is a defect, and it gets named in the reply (what ran too long, where it was moved), never buried.
The doors
| what it does | door |
|---|---|
| the map + spec as one page (what the sidebar renders) | npx thetacog-mcp steer-ui |
| the ballistic walk on the chip — the three panels | npx thetacog-mcp cockpit |
| the shape, its movement, its meaning | npx thetacog-mcp story |
| nine rooms vs declared poles | npx thetacog-mcp alignment-map |
| the operator's shape | npx thetacog-mcp competence-pixel |
| the trigger predicate + backtest | npx thetacog-mcp parametric-trigger |
| both arms + the pull | npx thetacog-mcp envelope |
| the net — strands, holes, mesh size | npx thetacog-mcp mesh |
| the Merkle spec tree | npx thetacog-mcp spec-tree |
| the goal — status · copy · import | npx thetacog-mcp goal |
| the one loop door — a verb, or free text | npx thetacog-mcp steer <args> |
| compile a dictation into ONE steer txt and RUN it (Template M per ask, the six percentages, the one-line goal; the steer starts at once) | npx thetacog-mcp steer-compile "<dictation>" · --check |
| workers — list · end one · end all | npx thetacog-mcp workers <args> |
| keep the loop running — on · off · status | npx thetacog-mcp keep-running <args> |
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.
- yesterday Changed · +4 lines 593fa30d30b1
- 8d ago First seen · 94 lines · 126 tokens per session scan A 18cbe628df45
steer is a skill published in the GitHub repository wiber/thetacog-mcp (1 stars, last pushed yesterday), licensed MIT. It adds 126 tokens to every session and 1,832 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-10-30.
steer compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| steerwiber/thetacog-mcp | Skill | 1 | yesterday | MIT | A | 126 | Claude Code, Codex |
| git-checkpointagentscope-ai/agentscope-java | Skill | 6.3k | today | Apache-2.0 | A | 39 | Claude Code, Codex |
| synthesize-reportagentscope-ai/agentscope-java | Skill | 6.3k | today | Apache-2.0 | A | 29 | Claude Code, Codex |
| data-analysisagentscope-ai/agentscope-java | Skill | 6.3k | today | Apache-2.0 | A | 25 | Claude Code, Codex |
| summarizeragentscope-ai/agentscope-java | Skill | 6.3k | today | Apache-2.0 | A | 24 | Claude Code, Codex |
| agent-browserTencent/WeKnora | Skill | 37k | 4d ago | none | A | 83 | Claude Code, Codex |
| skill-creator-primersammcj/agentic-coding | Skill | 169 | 3d ago | Apache-2.0 | A | 74 | Claude Code |