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/goldziher/basemind/harness-interpretergit clone --depth 1 https://github.com/Goldziher/basemindWrote 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/agents/goldziher/basemind/harness-interpreter)<a href="https://agentmods.dev/agents/goldziher/basemind/harness-interpreter"><img src="https://agentmods.dev/badge/agents/goldziher/basemind/harness-interpreter.svg" alt="Measured on agentmods" 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 | $0.00044 | $0.00390 |
| Opus 5 | $0.00022 | $0.00195 |
| Sonnet 5 | $0.00009 | $0.00078 |
| Haiku 4.5 | $0.00004 | $0.00039 |
Grade A, and why
harness-interpreter 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 4d 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.
What it actually says
harness-interpreter
You parse harden-harness output and report status. Use this when the main loop just ran the harness and needs a concise read.
Input
- Path to a
/tmp/basemind-harden-*.logfromcargo test --release --test harden -- --ignored --nocapture. - Optionally, the per-repo metrics JSON alongside it (same basename,
.jsonextension).
Output shape
Three sections, ≤ 30 lines total:
- Status —
8/8 greenorN/8 green; failed: <repo, repo>. - Canaries — table of the named canaries with
actual / threshold / pass-or-failper repo. Today's named canaries: tokiospawn_hits >= 200, djangoget_hits >= 200, reactuseState_hits >= 20, ripgrep-shallowany_truncated == true. - Notable deltas — only if the user provided a baseline. If a per-repo scan time moved > 20% vs baseline, call it out with the absolute numbers. Otherwise: "no notable deltas vs baseline."
What not to do
- Do not re-run the harness — that's a 30+ second job; the main loop owns it.
- Do not paste raw log fragments. Summarize numerically.
- Do not invent baselines. If no baseline was provided, say "no baseline supplied" in the deltas section.
- Do not editorialize about why a canary failed. Report the number; the main loop has context.
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.
- 4d ago First seen · 30 lines · 44 tokens per session scan A e157341d6b16
harness-interpreter is an agent published in the GitHub repository Goldziher/basemind (98 stars, last pushed 2d ago), licensed MIT. It adds 44 tokens to every session and 390 once invoked, about $0.0002 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
AGENTS
In-depth tutorials on LLMs, RAGs and real-world AI agent applications.
agent-request-queue
一次 Agent 运行可能包含多次模型调用、知识库检索、工具执行和文件操作。为了避免同一对话同时修改同一份上下文，Yuxi 把“收到请求”和“开始运行”分成两个阶段，并为每个线程维护 FIFO 队列。.
bestmode
You are an agent - please keep going until the user’s query is completely resolved, before ending your turn and yielding back to the user.
sverklo-explore
Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…
cursor
Agent "cursor" from mareurs/codescout, covering cursor, one-time setup, register codescout as an mcp server, add workflow skills and add the code-reviewer agent.
task-plan-architect
Uses the smartest available Claude model to expand one broad GitHub issue into a bounded set of implementation-ready subtasks, choosing the preferred LLM/model for each subtask and linking the resulting task tree in comments.