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/wyattowalsh/agentsnpx agentmods add skills/wyattowalsh/agents/harness-masterWrote 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/wyattowalsh/agents/harness-master)<a href="https://agentmods.dev/skills/wyattowalsh/agents/harness-master"><img src="https://agentmods.dev/badge/skills/wyattowalsh/agents/harness-master/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/wyattowalsh/agents/harness-master"><img src="https://agentmods.dev/badge/skills/wyattowalsh/agents/harness-master.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 17 findings, up to medium
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 →
- medium MCP Rug Pull · line 34 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 114 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 192 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 284 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 348 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 400 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium Prompt Injection · line 23 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 26 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 36 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 44 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Rogue Agent · line 102 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
- medium Agent Snooping · line 157 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 Prompt Injection · line 251 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 253 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 341 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 369 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- low Excessive Agency · line 191 Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.Fix: Limit the skill's scope to its documented purpose. Remove instructions that enable the agent to perform actions outside its stated functionality.
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 | $0.00054 | $0.06983 |
| Opus 5.5 | $0.00022 | $0.02793 |
| Sonnet 5.5 | $0.00011 | $0.01397 |
| Haiku 4.5 | $0.00005 | $0.00698 |
Grade A, and why
harness-master scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 418 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harness Master
Audit AI harness configuration quality, usage posture, and ecosystem options, then apply approved config fixes in the same session.
NOT for: creating agents (agent-conventions), building MCP servers (mcp-creator), generic application telemetry (observability-advisor), or generic code review without harness/config focus (review).
Input: harness names, level, or natural language. Modes: Audit, Discover (gap + bounded research), Usage, Apply/Install, Journal. Run scripts/discovery/classify_intent.py --args "$ARGUMENTS" --json when mode is ambiguous.
Dispatch
$ARGUMENTS |
Mode | First move |
|---|---|---|
Empty |
Intake | Ask for harnesses or all, then ask for project, global, or both |
all |
Intake | Ask only for level |
all <level> / <level> all |
Audit | Dry-run all supported harnesses in deterministic order |
<harness...> |
Intake | Resolve one or more harnesses, then ask only for missing level |
<level> |
Intake | Resolve level, then ask only for missing harnesses |
<harness...> <level> / <level> <harness...> |
Audit | Run dry-run review for the selected harnesses and level |
grok-build both in a repo with the rig workflow capability |
Audit + receipt | Audit plugin ownership/collisions and emit a hash-bound saved-workflow projection preview; do not apply |
apply approved |
Apply Approved | Apply the last matching dry-run batch only if the user already approved it in this session |
apply <harness...> <level> / apply <level> <harness...> |
Apply Approved | Apply only if the same scope was already reviewed and approved in this session; otherwise rerun audit first |
install <harness...> |
Install Guidance | Show exact npx skills add ... commands only; do not edit configs |
discover |
Discover full | W0 scans → scouts → ideate → report per references/discovery-pipeline.md |
discover audit / audit gaps |
Discover W0 | Gap report only; no scouts |
discover resume / discover list |
Journal | Load or list ~/.agents/harness-master/discovery/ sessions |
usage <harness|all> [level] [days] |
Usage Review | Collect safe aggregate usage/cost/token/tool signals; read-only |
| NL: missing skills / expand / find best X for harness | Discover (inferred) | Depth: full, focused, candidate, or compare — see discovery-pipeline.md |
| Natural language: "review/audit/check/tune config" | Audit | Normalize harnesses + level, then run dry-run review |
| Natural language: "review/analyze usage/cost/tokens/tools for " | Usage Review | Normalize harnesses + level, collect safe metadata only, and report optimization lanes |
| Natural language: "find/latest/best plugins/extensions/MCPs/skills for " | Discover focused | W0 + category scouts + source_probe.py; read-only |
| Natural language approval like "approved", "do it", "apply those changes" | Apply Approved | Continue only if the immediately preceding harness-master review matches the scope and is still current |
| Requests to create agents or MCP servers | Refuse + redirect | Redirect to the correct specialized skill |
| Unsupported harness names | Refuse + clarify | List supported harnesses and ask the user to choose from that set |
What ships with it
60 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.
- data/discovery/agent-targets.json 152 B
- data/discovery/discovery-taxonomy.json 4.2 KB
- data/discovery/fixtures/hook-scan-minimal.json 4.5 KB
- data/discovery/schemas/gap-report.schema.json 1.4 KB
- data/discovery/schemas/hook-scan.schema.json 2.6 KB
- data/discovery/schemas/scout-artifact.schema.json 635 B
- data/discovery/schemas/wave-manifest.schema.json 1.6 KB
- data/research-sources.json 17 KB
- evals/absent-surfaces.json 461 B
- evals/all-both.json 361 B
- evals/apply-approved.json 482 B
- evals/apply-research-without-audit.json 585 B
- evals/apply-without-review.json 346 B
- evals/chatgpt-global.json 392 B
- evals/cherry-studio-global-expanded.json 392 B
- evals/claude-desktop-global.json 411 B
- evals/community-source-negative.json 625 B
- evals/credentialed-source-degraded.json 512 B
- evals/cursor-agent-alias.json 435 B
- evals/cursor-cli-both.json 596 B
- evals/cursor-cli-install.json 409 B
- evals/cursor-project.json 404 B
- evals/degraded-docs.json 496 B
- evals/empty-invocation.json 396 B
- evals/evals.json 15 KB
- evals/github-graphql-enrichment.json 719 B
- evals/harness-only.json 296 B
- evals/implicit-trigger.json 574 B
- evals/install-guidance.json 460 B
- evals/level-only.json 276 B
- evals/mcp-registry-cross-source.json 806 B
- evals/multi-harness-mixed-order.json 340 B
- evals/negative-control.json 386 B
- evals/perplexity-desktop-both.json 398 B
- evals/skills-source-audit-route.json 672 B
- evals/ui-blind-spots-negative.json 480 B
- evals/unsupported-harness.json 332 B
- evals/usage-review-cloud-blind-spots.json 564 B
- evals/usage-review-natural-language.json 570 B
- evals/usage-review-not-observability-advisor.json 435 B
- evals/usage-review-opencode-explicit.json 569 B
- evals/usage-review-raw-transcript-negative.json 611 B
- evals/usage-review-read-only-boundary.json 467 B
- evals/usage-review-recommendation-lanes.json 509 B
- evals/usage-review-secret-redaction.json 535 B
- references/discovery-pipeline.md 3.7 KB
- references/discovery/coordinator-contract.md 2.8 KB
- references/discovery/gap-analysis.md 1.4 KB
- references/discovery/output-formats.md 4.6 KB
- references/discovery/research-integration.md 912 B
- references/discovery/research-queries.md 4.5 KB
- references/discovery/scout-templates.md 3.4 KB
- references/discovery/team-templates.md 11 KB
- references/ecosystem-research.md 4.3 KB
- references/evidence-boundaries.md 5.0 KB
- references/harness-checklists.md 6.0 KB
- references/harness-surfaces.md 15 KB
- references/install-guidance.md 2.5 KB
- references/latest-doc-sources.md 6.1 KB
- references/output-format.md 6.9 KB
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 · 418 lines · 54 tokens per session scan A c77b45ac8c3d
harness-master is a skill published in the GitHub repository wyattowalsh/agents (5 stars, last pushed 1mo ago), licensed MIT. It adds 54 tokens to every session and 6,983 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…