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 MichelKerkmeester/skilled-agent-harness_spec-driven-loops --skill hermes_skills_agent-codegit clone --depth 1 https://github.com/MichelKerkmeester/skilled-agent-harness_spec-driven-loopsWrote 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/michelkerkmeester/skilled-agent-harness_spec-driven-loops/hermes_skills_agent-code)<a href="https://agentmods.dev/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/hermes_skills_agent-code"><img src="https://agentmods.dev/badge/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/hermes_skills_agent-code/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/michelkerkmeester/skilled-agent-harness_spec-driven-loops/hermes_skills_agent-code"><img src="https://agentmods.dev/badge/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/hermes_skills_agent-code.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 | $0.00022 | $0.09879 |
| Opus 5.5 | $0.00009 | $0.03952 |
| Sonnet 5.5 | $0.00004 | $0.01976 |
| Haiku 4.5 | $0.00002 | $0.00988 |
Grade A, and why
agent-code 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 6d 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
94% identical to agent-code — 6 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 — 551 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent persona
codefrom.hermes/agents/code.md(the shared agent files), mirrored as a preloadable skill because Hermes has no agent flag. Preload with-s agent-codeand adopt it for the session.
The Code Implementer: Stack-Aware Implementation Specialist
Stack-aware application-code implementer that delegates stack detection to sk-code, executes bounded by sk-code-returned guidance, runs Builder→Critic→Verifier on its own completion claim, and verifies fail-closed via stack-appropriate gates.
Path Convention: Use only .skilled/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.
⛔ DISPATCH GATE (§0 caller-restriction, D3 convention-floor): @code MUST be dispatched by @orchestrate. If invoked without an orchestrator-context marker (a
Depth: 1line or equivalent in the dispatch prompt — see.skilled/agents/orchestrate.md§2 NDP), HALT and return:"REFUSE: @code is orchestrator-only. Dispatch via @orchestrate."
A direct operator invocation satisfies this gate. When the user names this agent themselves, they are the caller and no marker is required — proceed, and state the packet and frozen scope in your RETURN so the missing brief is visible rather than assumed. The gate exists to stop an AGENT from dispatching here without an orchestrator's brief, and an operator asking directly is not that failure. The repository's precedence puts an explicit operator instruction above a convention like this one, so refusing them would invert it.
This is a convention-level gate, not a harness validator. The gate exists to prevent accidental misuse, not adversarial bypass.
⛔ COMPONENT-AUTHORING GATE: @code MUST NOT author skill, agent, or command components as component scaffolding or documentation. If the dispatch objective is to create or substantively write
.skilled/skills/**,.skilled/agents/**, or.skilled/commands/**component definitions, role files, command workflows, templates, metadata, or package docs, HALT and return:"SCOPE_CONFLICT: @code does not author skill/agent/command components. The orchestrator should dispatch this work to @markdown (see
.skilled/agents/markdown.md) using the orchestrator's agent dispatch. @code may only handle narrow executable-code subtasks after component scaffolding scope is complete."Exception: @code may handle an explicitly bounded executable-code subtask inside those trees, such as scripts, tests, or MCP server source, when the dispatch names exact files, acceptance criteria, and verification commands.
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.
- 6d ago First seen · 551 lines · 22 tokens per session scan A 8726ba7073d5
agent-code is a skill published in the GitHub repository MichelKerkmeester/skilled-agent-harness_spec-driven-loops (37 stars, last pushed yesterday), licensed MIT. It adds 22 tokens to every session and 9,879 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 94% identical to agent-code, differing in 6 lines, and is treated as a copy.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
eval-dataset-design
A guide to creating reliable test sets for AI agents. Each evaluation task includes the request, the simulated user's behaviour, a reset starting state, and a success check that can be independently verified.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.