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 berekvolgyipeter/dotclaude --skill agent-harnessgit clone --depth 1 https://github.com/berekvolgyipeter/dotclaudeWrote 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/berekvolgyipeter/dotclaude/agent-harness)<a href="https://agentmods.dev/skills/berekvolgyipeter/dotclaude/agent-harness"><img src="https://agentmods.dev/badge/skills/berekvolgyipeter/dotclaude/agent-harness/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/berekvolgyipeter/dotclaude/agent-harness"><img src="https://agentmods.dev/badge/skills/berekvolgyipeter/dotclaude/agent-harness.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 | $0.00129 | $0.00785 |
| Opus 5.5 | $0.00052 | $0.00314 |
| Sonnet 5.5 | $0.00026 | $0.00157 |
| Haiku 4.5 | $0.00013 | $0.00078 |
Grade A, and why
agent-harness 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 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.
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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Harness
Help users discover, explore, and understand agent harness frameworks and Claude Code resource collections from curated repositories.
Catalog
The catalog at ~/.claude/skills/agent-harness/references/catalog.json lists all indexed repositories with descriptions and topics. Read it first to understand what's available and which repo is most relevant to the user's need.
Each entry has these fields: github (source repo), local_path (where it's cloned), description (what it contains), topics (keyword tags), and content_types (e.g. agents, frameworks, resources).
All repos are cloned and indexed together under a single path:
~/.claude/skills-references/agent-harness/
Each repo lives at the local_path specified in the catalog (e.g., ~/.claude/skills-references/agent-harness/coleam00-archon/).
Search
Always start with semantic search — it combines semantic understanding with BM25 keyword matching, so it surfaces conceptually relevant results even when exact keywords differ. Only fall back to Grep if semantic search returns no results or the index is unavailable.
Primary — semantic search:
mcp__claude-context__search_code(
path="~/.claude/skills-references/agent-harness",
query="<user need>"
)
Fallback — keyword search (only when semantic search fails or is unavailable):
Grep(pattern="<keyword>", path="~/.claude/skills-references/agent-harness/", glob="*.md")
Browse
To list contents of a specific repo:
Glob(~/.claude/skills-references/agent-harness/<repo-dir>/**/*.md)
Check for structured metadata files if present:
Glob(~/.claude/skills-references/agent-harness/<repo-dir>/**/*manifest*.json)
Glob(~/.claude/skills-references/agent-harness/<repo-dir>/**/*config*.json)
Workflow
- Understand the need — ask clarifying questions if the user's request is vague (e.g., "agent" could mean orchestration framework, a specific Claude Code agent, or a resource collection).
- Catalog — read
~/.claude/skills/agent-harness/references/catalog.jsonto identify which repos are relevant based on descriptions and topics. - Search — use semantic search to find matching content across all repos.
- Read before recommending — read the relevant README, docs, or source files before presenting findings. Understand what it does, what dependencies it needs, and how to use it.
- Present matches — show name, description, source repo, and category. When multiple options exist, compare them briefly.
- Guide usage — explain how to run, configure, or integrate what was found. Point to setup docs, env vars, or prerequisites as needed.
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 71 lines · 129 tokens per session scan A 421e5f74bb5b
agent-harness is a skill published in the GitHub repository berekvolgyipeter/dotclaude (2 stars, last pushed today), licensed MIT. It adds 129 tokens to every session and 785 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 2026-08-31.
Other skills, from other repositories
accint-solve
Route agent work through AccInt's MCP memory loop: retrieve prior outcomes, resolve frames, and close commitments with evidence.
brainstorming
Interactive spec generation — turn ideas into concrete specs with R-numbered requirements and testable acceptance criteria.
collaborating
Hackathon-native multiplayer collaboration mode -- brain-dump together, claim tasks across machines, flag decisions async. Drives the subcommands of /forge:collaborate.
executing
Autonomous task implementation workflow — implement, test, review, commit for each task in the frontier.
backpropagation
Trace runtime bugs back to spec gaps — identify missing acceptance criteria, update specs, generate regression tests, and detect patterns.
caveman-internal
Internal token optimization for Forge agent artifacts (handoff notes, artifact summaries, review notes). NOT exposed as a user-facing /caveman command.