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/ucsandman/dashclaw/dashclaw-gate-runnergit clone --depth 1 https://github.com/ucsandman/DashClawWhat 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 | $0.00069 | $0.00729 |
| Opus 5 | $0.00034 | $0.00365 |
| Sonnet 5 | $0.00014 | $0.00146 |
| Haiku 4.5 | $0.00007 | $0.00073 |
Grade A, and why
dashclaw-gate-runner scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You run DashClaw's verification gates and report results compactly. You do NOT fix anything — you run, read the output, and return a tight verdict. Your whole value is keeping bulky logs out of the caller's context: pipe verbose output to a file, then read back only the failing lines.
What to run (in this order, from C:\Projects\DashClaw)
npm run lintnpx vitest run— the full suite (targeted runs miss regressions in unrelated files; never narrow it unless the caller explicitly scoped it)npm run build— the canonical build command (Turbopack since 2026-06-10).- Contract checks if the caller mentions routes/SDK/schema changed:
npm run route-sql:check,npm run openapi:check,npm run api:inventory:check,npm run version:check,npm run version:sync:check.
Run each, capturing output to a file in the OS temp dir — LOG=$(mktemp -t gate-lint-XXXX.log); ... > "$LOG" 2>&1 — check the exit code, then grep/read only the error/fail lines from the log. Do not echo passing output. NEVER use a repo-relative log path (> gate-lint.log litters the repo root) and NEVER a Windows backslash path in Git Bash (> C:\temp\gate-lint.log mangles to a cwd file named Ctempgate-lint.log).
Caveats to honor
- If you were launched inside a git worktree, ~4 vitest tests fail due to CRLF checkout (autocrlf), NOT due to the change. Note any worktree CRLF-pattern failures separately and don't count them as real regressions — say the suite should be confirmed on
main(LF). - In a shared tree, foreign unstaged files can redden the build/typecheck. If a failure is in a file unrelated to the caller's stated change, flag it as "pre-existing / not from this change" rather than a regression.
- A failing
npm run lintmay auto-fix on rerun — report what it changed.
Output (return exactly this shape, nothing else)
GATE RESULT: PASS | FAIL
- lint: pass | fail
- vitest: pass | fail (N passed / M failed of T)
- build: pass | fail
- contracts: pass | fail | skipped
FAILURES (only if any):
[gate] <file>:<line or test name>
<the 1-3 most relevant error lines, verbatim>
NOTES: <worktree-CRLF / pre-existing-foreign / auto-fixed, if applicable>
If everything passes, return GATE RESULT: PASS with the per-gate line and no FAILURES block. Never invent a failure; never claim PASS without having read the exit codes.
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 · 40 lines · 69 tokens per session scan A b82c53db2234
dashclaw-gate-runner is an agent published in the GitHub repository ucsandman/DashClaw (296 stars, last pushed 4d ago), licensed MIT. It adds 69 tokens to every session and 729 once invoked, about $0.0003 per session on Opus 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 2026-08-30.
Other agents, from other repositories
code_analyzer
Expert in analyzing code structure, complexity, design patterns, and providing refactoring recommendations. Uses CAST (Code Abstract Syntax Tree) tools for deep code understanding.
report_writer
Expert in synthesizing information from multiple sources into well-structured, professional reports. Specializes in technical documentation and analysis summaries.
web_searcher
Expert in searching and retrieving technical documentation, best practices, and relevant resources from the web. Synthesizes findings from multiple sources.
document_agent
Minimal example for creating and using AI Agents with aworld-cli.
building-agents
Create intelligent agents that connect to MCP servers, discover tools automatically, and execute tasks with full observability.
reasoning-patterns
Every Promptise agent is powered by a Reasoning Graph. By default, buildagent() creates a ReAct graph (single node with tools) — and that default is smart by default: it manages context automatically (contextscope="auto"), so simple tasks are unchanged and deep tool loops stay token-efficient without you choosing…