Borrowing it
Nothing to install: this file belongs to Sandeeprdy1729/timps-swarm. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Sandeeprdy1729/timps-swarm/main/AGENTS.mdgit clone --depth 1 https://github.com/Sandeeprdy1729/timps-swarmWrote 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/instructions/sandeeprdy1729/timps-swarm/agents-md)<a href="https://agentmods.dev/instructions/sandeeprdy1729/timps-swarm/agents-md"><img src="https://agentmods.dev/badge/instructions/sandeeprdy1729/timps-swarm/agents-md.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 5.1 | $0.03150 | $0.03150 |
| Opus 5 | $0.01575 | $0.01575 |
| Sonnet 5 | $0.00630 | $0.00630 |
| Haiku 4.5 | $0.00315 | $0.00315 |
Grade B, and why
timps-swarm AGENTS.md scanned grade B with 2 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- `TIMPS_AUTH=1` enables API key auth; keys are SHA-256 hashed in `~/.timps/.secrets` (chmod 600). Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Health: `curl http://localhost:8000/health` (and `/health/full` for provider status). Interactive docs at `/docs`. How it starts
The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for OpenCode sessions working in the TIMPS Swarm monorepo. Only signals you would miss from a quick read of the README are listed here.
Layout
src/main.py— FastAPI server (port 8000). Lazy-imports the LangGraph_graphto keep tests/imports cheap. Also exposesGET /mcp/toolsandPOST /mcp/tools/callso the Node.js MCP proxy can reach every tool over HTTP.src/agents.py,src/more_agents.py,src/expert_agents.py,src/developer_swarm_agents.py,src/intelligence_agents.py,src/aiml_agents.py,src/knowledge_swarm_agents.py,src/computer_agents.py,src/priority_agents.py— flat agent modules.src/aiml/,src/business/,src/domain/,src/emerging/,src/india/,src/infra/,src/intelligence/,src/nextgen/,src/phase7/— subpackages of specialist agents. Each is a sibling flat namespace; import asfrom src.nextgen.foo import foo.src/layer1_computer_manager.py,src/layer2_swarm_bridge.py,src/layer3_swarm_cli.py— the 3-layer runtime (per-agent compute isolation, swarm bridge/orchestration, CLI shim). Spawned agents supporttask_idfor budget tracking.src/observer.py— TaskObserver: sliding-window call buffer, args fingerprinting, repetitive-pattern detection; firesspawn_callbackswhen a tool is invoked ≥3 times with similar args.src/auto_spawner.py— AutoSpawner: wired to observer callbacks, maps MCP tool names → AgentRole, spawns sub-agent batches vialayer2_swarm_bridge.src/context_sharing.py— ContextSharer: captures git status, recent diff, file tree, framework detection, test examples, style config, and open editor files for injection into sub-agent contexts.src/decomposer.py— TaskDecomposer: decomposes a high-level instruction into concrete sub-tasks by filesystem scan (keyword → glob) or LLM fallback. Powers thetimps_batchtool.src/delegation_engine.py— DelegationEngine: orchestrates the full batch-delegation pipeline (context → decompose → dependency waves → parallel execute → normalize → aggregate). IncludesResultNormalizerandResultAggregatorfor merging heterogeneous agent outputs.src/failure_recovery.py— FailureRecovery: progressive retry with prompt modification (simplify → rephrase → minimal). Maps error types to strategies; max 3 retries.src/token_budget.py— TokenBudgetManager: per-task ($0.50), per-run ($10), daily ($50) cost tracking with hard limits. Injected intollm_router.pyto record every LLM call.src/llm_router.pyis a thin façade — provider logic lives insrc/providers/*.py(gemini, anthropic, openai, groq, ollama, mcp_sampler, timps, null). All providers userequests.Sessionwith connection pooling (HTTPAdapter pool_maxsize=50) and retry-on-status (429, 5xx). LLM call costs are tracked viaTokenBudgetManager.mcp_server/server.py— MCP stdio server (JSON-RPC 2.0). Uses aThreadPoolExecutor(max_workers=10)to run tool handlers in background threads, sendingnotifications/progressmessages to the client. Registers an MCP-sampling callback that the LLM router prefers over cloud providers when the host supports it. DefinesTOOLS(the 161-tool catalogue) and_TOOL_HANDLERS(name → lambda). The FastAPI server lazy-imports both for/mcp/toolsand/mcp/tools/call. Exposestimps_batchfor intelligent bulk delegation.cli/— the published npm package (@timps-swarm/cli-style). Layout:bin/timps-swarm.js,lib/local-audit.js,lib/agents-manifest.js(161 agent metadata),lib/agent-files.js(sub-agent.mdwriter/remover),lib/mcp-proxy.js(Node.js JSON-RPC 2.0 stdio proxy formcpwhen Python repo missing),tools/generate-manifest.js(Python-AST re-sync). Has its ownpackage.json,node_modules/,package-lock.json.postinstallprints thenpx timps-swarm install-mcpnext step; running that command writes 1) thetimps-swarmMCP server entry into supported IDE config files (withenv:block forwardingANTHROPIC_API_KEY,OPENAI_API_KEY,GEMINI_API_KEY,GROQ_API_KEY,TIMPS_API_URL,OLLAMA_HOST,REDIS_URL) and 2) one sub-agent.mdfile per TIMPS tool into~/.claude/agents/,./.claude/agents/, and~/.codex/agents/.timps-swarm uninstall-mcpreverses both. Add--no-sub-agentsto install-mcp to skip the.mdfiles.dashboard/— React UI (react-scripts). Talks to the API viaproxy: http://localhost:8000and a WebSocket atws://localhost:8000/ws.- Top-level Python entry points:
give_work.py(SDLC/health dispatch),timps_cli.py(unifiedtimpsCLI),timps_daemon.py(context-keeper daemon),timps_doctor.py(env diagnostics). generated/— runtime artefacts (created on first run; mounted as/app/generatedin Docker).dist/— built wheel/sdist frompyproject.toml(do not edit by hand).
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 · 85 lines · 3,150 tokens per session scan B b94b1c1a4bc5
timps-swarm AGENTS.md is an instructions file published in the GitHub repository Sandeeprdy1729/timps-swarm (1 stars, last pushed 5d ago), licensed MIT. It adds 3,150 tokens to every session, about $0.0158 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.