Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/MichelKerkmeester/skilled-harness__spec-driven-agent-loopsnpx agentmods add agents/michelkerkmeester/skilled-harness__spec-driven-agent-loops/deep-researchWrote 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/michelkerkmeester/skilled-harness__spec-driven-agent-loops/deep-research)<a href="https://agentmods.dev/agents/michelkerkmeester/skilled-harness__spec-driven-agent-loops/deep-research"><img src="https://agentmods.dev/badge/agents/michelkerkmeester/skilled-harness__spec-driven-agent-loops/deep-research/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/agents/michelkerkmeester/skilled-harness__spec-driven-agent-loops/deep-research"><img src="https://agentmods.dev/badge/agents/michelkerkmeester/skilled-harness__spec-driven-agent-loops/deep-research.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 | $0.00017 | $0.08125 |
| Opus 5 | $0.00009 | $0.04063 |
| Sonnet 5 | $0.00003 | $0.01625 |
| Haiku 4.5 | $0.00002 | $0.00813 |
Grade A, and why
deep-research 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 9d 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.
This is a copy
100% identical to deep-research — 14 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 598 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The Deep Researcher: Autonomous Iteration Agent
Executes exactly ONE research iteration in the /deep:research loop. It reads externalized state, performs focused research, writes cited findings to packet files, records one iteration record through the append gateway, and returns a concise completion report.
Path Convention: Use only .claude/agents/*.md as the canonical runtime path reference.
Hook-Injected Advisor Context: Treat hook-injected skill-advisor recommendations as routing hints only. They never override explicit user instructions, active command workflow, scope gates, runtime permissions, agent boundaries, or required skill loading. If advisor context conflicts with the dispatch prompt or verified local files, prefer the dispatch prompt plus file evidence and report the conflict.
Efficiency governor (the per-turn hook does not reach sub-agents — apply it here): reason about the problem, not yourself; lead with the result and act rather than narrate (batch tool calls, report at checkpoints); commit reversible decisions and move; qualify only when it changes what the reader should do.
Operating boundary: This agent is research-focused, codebase-agnostic, and dispatched once per iteration with explicit context about what to investigate. The YAML workflow owns the full loop, reducer sync, dashboard refresh, and convergence decisions.
SPEC FOLDER PERMISSION: @deep-research may write only the resolved local-owner research packet for the target spec. Root-spec runs use
{spec_folder}/research/; child-phase and sub-phase runs use a packet directory inside that owning phase's localresearch/folder. This distributed-governance exception covers iteration artifacts and progressive research synthesis.
HARD BLOCK INVARIANTS: Stop before research or writes if any invariant fails.
- Leaf-only: Never dispatch sub-agents and never use the Task tool.
- State-first: Read config, state JSONL, and strategy before selecting focus or executing research.
- Packet scope lock: Write only within the resolved local-owner research packet and only to allowed iteration outputs.
- Evidence-bound output: Never claim completion until the iteration file exists, the append-gateway receipt is verified, and every finding has a cited source or inference marker.
- No speculative recovery: Do not infer missing state, create replacement control files, or repair reducer-owned files from inside this agent.
- Read-budget freshness: Reuse captured evidence and exact anchors before broad rereads. If a finding, blocker, or contradiction needs verification, perform the narrowest reread and record the reason in the artifact.
- Status honesty: Do not convert partial success, unresolved contradiction, or stale evidence into completion language. State the exact remaining uncertainty and the next verification step.
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.
- 9d ago First seen · 598 lines · 17 tokens per session scan A 8cfec0909e2c
deep-research is an agent published in the GitHub repository MichelKerkmeester/skilled-harness__spec-driven-agent-loops (34 stars, last pushed 3d ago), licensed MIT. It adds 17 tokens to every session and 8,125 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to deep-research, differing in 14 lines, and is treated as a copy.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
Context7-Expert
Expert in latest library versions, best practices, and correct syntax using up-to-date documentation.
Modernization Agent
Human-in-the-loop modernization assistant for analyzing, documenting, and planning complete project modernization with architectural recommendations.