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 instructions/paultyng/testagent/agents-mdgit clone --depth 1 https://github.com/paultyng/testagentWhat 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 | $0.02165 | $0.02165 |
| Opus 5 | $0.01082 | $0.01082 |
| Sonnet 5 | $0.00433 | $0.00433 |
| Haiku 4.5 | $0.00216 | $0.00216 |
Grade A, and why
testagent AGENTS.md 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 — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Context for AI agents working on this repo. (Cursor, Claude Code, and others read this file.)
What is testagent?
A fake CLI agent for testing orchestration tooling that drives real coding agents (Claude Code, Codex, Gemini, GitHub Copilot CLI, etc.). Runs as an interactive PTY process; emits the same kinds of terminal artifacts (boxed tool-use blocks, streaming text, in-place updates) and the same protocol traffic (HTTP hooks, MCP JSON-RPC) as a real agent — without calling any LLM.
The product framing is deterministic output for tests, not "no LLM" as a virtue. Wiring a local LLM into a test harness is a valid choice for some workflows; testagent is explicitly not that. Every assertion-relevant byte (slash dispatch, hook payloads, MCP frames, stream-json shapes) is scripted by the user, not generated, so tests stay stable across runs.
v1 ships a drop-in fake for Claude Code: argv compatibility for the flags orchestrators commonly emit, HTTP and command hooks (UserPromptSubmit / PreToolUse / PostToolUse / Stop / SessionStart / SessionEnd / PreCompact / PostCompact), an MCP HTTP client that handshakes and dispatches tools/call, --print --output-format stream-json for non-interactive callers, a slash-command grammar for driving UI primitives interactively (including /clear and /compact which fire the corresponding hook lifecycles without restarting the process), and lipgloss-rendered plausible-shape output.
Future phases
| Phase | What |
|---|---|
| 3+ | Vendor-neutral refactor; per-agent-type subcommands + packages (see Future conventions); adapters for codex, gemini, copilot, plus tier-2 (aider, amp, q, goose, crush); conformance tests against captured real-vendor traffic |
| release | goreleaser cross-OS binaries (darwin/linux/windows × amd64/arm64) + signed releases |
Design conventions
- Schema types are duplicated, not imported.
SettingsandMCPConfigmirror Claude Code's on-disk shapes and live incmd/claude/settings.go(vendor-specific, not shared with the engine). - Stdlib-first; deps are deliberate. Each non-stdlib dep is justified in the commit message that adds it (
mark3labs/mcp-go,lipgloss,bubbletea,bubbles,go-isatty,spf13/cobra). - Interactive vs non-interactive split. TTY stdin → bubbletea TUI (
internal/engine/tui.go, inline rendering with native terminal scrollback, concurrent input during the thinking spinner). Piped stdin → scanner loop (internal/engine/scanner.go, line-based, inline rendering).--print/-pis a third path (cmd/claude/print.go, one-shot output formatter). Themattn/go-isattycheck onos.Stdinis the TUI/scanner gate; e2e tests pipe stdin so they always hit the scanner path. - Conventional Commits. One commit per phase. Each phase's commit leaves the tree buildable and tested.
- Tests:
t.Parallel(), table-driven, realhttptest/exec-driven integration over mocks where possible (seee2e_test.go). Fixtures intestdata/. Time-dependent helpers (anything that callstime.Sleep/time.NewTimer/tea.Tickoutside real network IO) usetesting/synctestso virtual time advances without real wall-clock waits — seeinternal/engine/{spinner,stream}_test.gofor the pattern. - Debug output goes to stderr. Verbose / debug logging (e.g.
--verbosehook traces) is plain text, one event per line, never ANSI-styled — it gets grepped and piped. Stdout stays reserved for stream-json frames and TUI rendering. COMPATIBILITY.mdis the contract artifact. It is the per-vendor source of record for which flags, slash commands, and REPL behaviors testagent implements, accepts, or omits. Update it when adding vendor-facing features. Use theupdate-compatibilityskill (one skill, dispatches by vendor —claude,codex, orcursor) to refresh a section after an upstream release or when drift is detected. Per-vendor source URLs, probing recipes, and matrix templates live under the skill'svendors/<name>/subdirectories. Cursor adapter is in early-stage planning — the template is ready; vendor wiring lands in a future phase.RELEASING.mdis the release-policy doc (semver scope, bump table, runbooks). When cutting a release, drive it through thepublish-releaseskill, which executes RELEASING.md step by step and gates on the two human decisions (version number, release-notes copy).
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 · 94 lines · 2,165 tokens per session scan A 64a998e412fe
testagent AGENTS.md is an instructions file published in the GitHub repository paultyng/testagent (7 stars, last pushed 1mo ago), licensed MIT. It adds 2,165 tokens to every session, about $0.0108 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 2026-08-31.
Other instructions, from other repositories
apm python.instructions.md
Python development guidelines.
CodexPotter AGENTS.md
Instructions for breezewish/CodexPotter, covering repository guidelines, workflow principles, engineering rules, core principles: simplicity & readability and better maintainability.
copilot-proxy-api CLAUDE.md
Instructions for voidsteed/copilot-proxy-api, covering claude.md, project overview, common commands, architecture and entry points & cli structure.
OpenMicro AGENTS.md
Instructions for stephenleo/OpenMicro, covering openmicro — project conventions and releases.
squarebox CLAUDE.md
Instructions for SquareWaveSystems/squarebox, covering claude.md, agent skills, issue tracker, triage labels and domain docs.
SubFrame AGENTS.md
Instructions for Codename-11/SubFrame, covering subframe - subframe project, core working principle, relationship to native ai tools, session start and concurrent work & worktrees.