Borrowing it
Nothing to install: this file belongs to friedbotstudio/baseline. 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/friedbotstudio/baseline/main/.claude/skills/harness/SKILL.mdgit clone --depth 1 https://github.com/friedbotstudio/baselineWrote 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/friedbotstudio/baseline/harness)<a href="https://agentmods.dev/skills/friedbotstudio/baseline/harness"><img src="https://agentmods.dev/badge/skills/friedbotstudio/baseline/harness/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/friedbotstudio/baseline/harness"><img src="https://agentmods.dev/badge/skills/friedbotstudio/baseline/harness.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 11 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Tool Misuse · line 60 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 113 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 117 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 145 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Agent Snooping · line 185 Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.Fix: Remove all code or instructions that access agent configuration directories (.claude/, .codex/, .gemini/). If configuration values are needed, pass them explicitly as parameters or environment variabl
- medium Excessive Agency · line 173 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 121 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 176 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 178 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Agent Snooping · line 177 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 178 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 | $0.00110 | $0.11259 |
| Opus 5.5 | $0.00044 | $0.04504 |
| Sonnet 5.5 | $0.00022 | $0.02252 |
| Haiku 4.5 | $0.00011 | $0.01126 |
Grade A, and why
harness 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 1mo 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 — 248 lines — stays where its author put it; the contents beside it link to each section on GitHub.
harness — workflow orchestrator with internal loop
User-invokable and model-invokable. The harness chains the 11-phase pipeline by looping internally through non-gated phases until the loop hits one of four exit conditions: consent gate, phase-skill failure, integrate-failure-needs-spec-change, or workflow done. The user types only at consent gates (/approve-direction, /approve-swarm, /grant-commit) and at integrate-failure decisions that need a spec change.
Internal loop atomicity (the contract)
A single Skill(harness) invocation loops through every non-gated phase boundary in one user turn. Inside the loop, each iteration invokes exactly one phase skill via the Skill tool, updates state and TaskList, then re-enters the loop. The loop exits — and the model emits its terminal message — only when one of these four conditions holds:
- Yield: the next pending task carries
metadata.needs_user: true(consent gate, or integrate-failure-needs-spec-change). Writeharness_state: yielded; surface the gate; exit. - Phase-skill failure: a
Skill(<phase>)call returned error. Writeharness_state: yieldedwithreason: "<phase> failed: <summary>"; surface; exit. - Done:
workflow.json → completednow contains every non-excepted phase. Writeharness_state: done; surface completion; exit. - (Rare) Mid-loop interruption: the model decides to stop emitting before any of the above (context pressure, runtime limit, external interruption). The on-disk state stays
state: continuewith the marker present — the Stop hook safety net handles this.
.claude/state/harness_state is flat JSON with one of four states:
continue— the harness is in the loop body (or was interrupted mid-loop). The Stop hook safety net is armed.yielded— the loop exited cleanly at a gate or failure. Stop hook stays silent.done— the loop exited cleanly at workflow completion. Stop hook stays silent.parked— a caller owns this session and the loop is not to be resumed. Stop hook stays silent, and stays silent even with the marker present, because a park happens inside an armed loop. Preflight step 6 clears it on the next/harness.
What ships with it
34 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- assemble-context.mjs 4.3 KB runs code
- changed-files-shape.mjs 1.5 KB runs code
- checker-fanout.mjs 10 KB runs code
- checkers/ac-conformance.mjs 2.2 KB runs code
- checkers/backlog-deferral.mjs 2.5 KB runs code
- checkers/mutation-score.mjs 2.8 KB runs code
- checkers/spec-lint.mjs 3.2 KB runs code
- checkers/spec-shippability.mjs 3.4 KB runs code
- cli.mjs 5.8 KB runs code
- codesign-reentry.mjs 758 B runs code
- consolidate-open-questions.mjs 5.8 KB runs code
- design-judge.mjs 4.7 KB runs code
- envelope.mjs 3.6 KB runs code
- evidence-ledger.mjs 2.4 KB runs code
- gate-collapse-resolver.mjs 1.5 KB runs code
- graduation-gate.mjs 2.4 KB runs code
- maker-checker.mjs 586 B runs code
- notify.mjs 13 KB runs code
- payload-estimate.mjs 2.5 KB runs code
- plan-diff.mjs 1.4 KB runs code
- plan-frame.mjs 1.1 KB runs code
- plan-store.mjs 8.5 KB runs code
- plan-wiring.mjs 2.1 KB runs code
- pre-implementation-gate.mjs 2.1 KB runs code
- proposal.mjs 3.3 KB runs code
- ralph-loop.mjs 3.3 KB runs code
- ratio.mjs 4.6 KB runs code
- reentry.mjs 2.6 KB runs code
- replan.mjs 3.5 KB runs code
- rightsize-gate.mjs 11 KB runs code
- timing-corpus.mjs 4.1 KB runs code
- verdict.mjs 1.4 KB runs code
- work-planner.mjs 4.3 KB runs code
- workflow-migrator.js 3.3 KB runs code
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.
- 1mo ago First seen · 248 lines · 110 tokens per session scan A 378e6fdc3bdb
harness is a skill published in the GitHub repository friedbotstudio/baseline (14 stars, last pushed 13d ago), licensed Apache-2.0. It adds 110 tokens to every session and 11,259 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
code-review-with-lsp
Code review with LSP-powered code intelligence. Uses MCP tools (diagnostics, hover, references, definition, symbols) for semantic code understanding, not just text grep.
i18n-check
An internationalization (i18n) check for software that supports multiple languages. It compares translation files with the text and translation keys used by the code.
vue-best-practices
A review checklist for Vue 2 and Vue 3 components, covering component names, prop declarations, TypeScript casts, API styles and complex template expressions.
gdpr-audit
GDPR compliance audit of code, plans, schemas, or IaC. Produces a findings report with article citations, severity, confidence, and fixes; flags evidence gaps. Not a substitute for a DPO/lawyer.
cognitive-walkthrough
Deep-dive usability evaluation of specific user tasks. Simulates novice user cognition step-by-step to identify learnability issues, unclear actions, and points of confusion.
owasp-ai-testing
AI trustworthiness testing using OWASP AI Testing Guide v1. Execute 32 test cases across 4 layers (Application, Model, Infrastructure, Data) with practical payloads and remediation.