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/auerbachb/claude-code-config/readmegit clone --depth 1 https://github.com/auerbachb/claude-code-configWhat 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 | $0.03562 |
| Opus 5 | $0.00000 | $0.01781 |
| Sonnet 5 | $0.00000 | $0.00712 |
| Haiku 4.5 | $0.00000 | $0.00356 |
Grade A, and why
README 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 — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Custom Agent Definitions
This directory contains custom agent definitions for the Phase A/B/C subagent workflow and PM task execution. Each agent is a .md file with frontmatter metadata and role-specific rules. The harness automatically injects the project CLAUDE.md hierarchy and .claude/rules/*.md into every custom subagent_type agent at spawn — no manual rule-file injection needed. Each definition embeds only role-specific workflow rules plus the SAFETY/MINDSET blocks as deliberate safety-critical restatements. Verification: .claude/reference/token-efficiency-audit-2026-07.md §FU-1.
How It Works
Claude Code's Agent tool supports a subagent_type parameter that references agent definition files in .claude/agents/. When spawning a subagent with subagent_type: "phase-a-fixer", Claude Code loads the agent whose frontmatter name: field matches that string — identity comes from name:, not the filename. The name: field is required; without it the agent file is not resolvable by subagent_type.
Claude Code scans .claude/agents/ at session start. A session restart is required for a newly added or edited agent file to be registered. The tools: key in frontmatter restricts which tools the agent may use; the deprecated allowed-tools: key is not recognized by the current schema.
Reaching these agents from other repos (issue #1189)
Definitions are discovered at two scopes: project (<repo>/.claude/agents/) and user (~/.claude/agents/), with project winning on a name: collision. Without the user-scope copy the phase agents exist only here, so /pm and /subagent lose their inline pipelines everywhere else — and lose them silently, which is the failure this leg closes.
setup-skills-worktree.sh Step 5b publishes each file as ~/.claude/agents/<name>.md → skills-worktree, mirroring the per-entry skills symlinks rather than the single directory symlink used for rules (the docs guarantee symlink-following for .claude/rules/ but say nothing about agents/; details in .claude/reference/portable-skill-resolution.md).
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 · 152 lines · 0 tokens per session scan A 5e42b5932683
README is an agent published in the GitHub repository auerbachb/claude-code-config (5 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,562 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
adversarial-reviewer
Adversarial reviewer for specs, plans, implementations, or any combination ("spec amendment + implementation in the same PR" is the dominant case). Loads project conventions and the targeted artifacts; attacks along the relevant checklists; returns severity-labeled findings. Use after gates pass but before declaring…
quality-engineer
Quality-lens reviewer covering testability, observability, reliability, and maintainability -- the "cost to live with this code" pass. Also drafts contract or construction tests on request. Reads effective repository guidance, the spec and plan if any, the diff, and nearby tests; flags test-shape problems (wrong…
review-all-shared-rules
Shared severity tiers, verification gate, quotas, and auto-drop rules included by every review-all agent prompt.
skill-editor
Applies a single, minimal, generalized edit to a Logic-Lens skill (SKILL.md / guide / shared file) given a concrete failure diagnosis. Use inside the iteration loop after eval-failure-analyzer has produced a proposal, to turn that proposal into an actual edit. Mutates files; does NOT run evals or sync the cache — it…
bugs-and-security
Scan changed code for logic bugs, security vulnerabilities (OWASP Top 10), completeness gaps, and error handling issues.
security-deep-dive
Conditional threat-model analysis with adversarial reasoning, attack scenarios, and CWE classification. Distinct from agent 02 (which does broad pattern scanning).