puppy-one-harness
Skill40 starsChanged 5 days agoApache-2.0View on GitHub
Use when running, grading, or extending the on-device model harness for Puppy One, including the review-queue judge handoff, capability probing, and the evolution ledger.
In this repository69 other files it gives its agents.
- AGENTS.mdAGENTS.md
- hushh-webapp/AGENTS.mdAGENTS.md
- impeccableSkill
- agent-orchestration-governanceSkill
- analytics-observability-governanceSkill
- animateSkill
- animation-vocabularySkill
- apple-designSkill
- autonomous-rca-governanceSkill
- backend-agents-operonsSkill
- backend-api-contractsSkill
- backend-runtime-governanceSkill
- backendSkill
- brandkitSkill
- client-env-paritySkill
- codex-skill-authoringSkill
- comms-communitySkill
- context-refreshSkill
- contributor-onboardingSkill
- docs-governanceSkill
- emil-design-engSkill
- find-animation-opportunitiesSkill
- founder-brief-curationSkill
- frontend-architectureSkill
- frontend-cache-coherenceSkill
- frontend-design-systemSkill
- frontend-native-surface-mapperSkill
- frontendSkill
- frontend-surface-placementSkill
- future-plannerSkill
- github-contribution-governanceSkill
- hushh-consent-mcpSkill
- hushh-icon-themeSkill
- iam-consent-governanceSkill
- imagegen-frontend-mobileSkill
- imagegen-frontend-webSkill
- improve-animationsSkill
- industrial-brutalist-uiSkill
- kai-test-account-resetSkill
- less-text-karma-jiSkill
- mcp-developer-surfaceSkill
- mobile-nativeSkill
- mobile-parity-auditSkill
- mobile-plugin-contractsSkill
- morphy-axSkill
- one-voice-governanceSkill
- oss-license-governanceSkill
- pdf-artifact-generationSkill
- pick-ui-librarySkill
- pkm-upgrade-rehearsalSkill
- planning-boardSkill
- pr-governance-reviewSkill
- profile-uiSkill
- prototypeSkill
- quality-contractsSkill
- release-ios-appstoreSkill
- repo-contextSkill
- repo-operationsSkill
- review-animationsSkill
- reviewer-app-testingSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 1 is distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
Avoid overlap with quality-contracts and backend-agents-operons.
- Very specific: line 63
Every wrong verdict cites the offending value verbatim.
- Distinctive: line 70
Before comparing two runs, check the capability profile.
- Very specific: line 75
If the request is still broad or ambiguous, route it back to backend.
- Very specific: line 76
If the finding is a provider-routing or gate issue, use backend-runtime-governance.
- Very specific: line 78
If it touches the vault or the PKM write path, use security-audit.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Puppy One Harness Skill
Purpose and Trigger
- Primary scope:
puppy-one-harness - Trigger on benchmarking a local model, grading its output, adding a model to the ladder, or reading the evolution ledger.
- Avoid overlap with
quality-contractsandbackend-agents-operons.
The harness answers two different questions and must never conflate them. The
benchmark asks well-formed and fast; the judge asks correct. A model can emit
a perfectly shaped save_to_pkm call that files a dietary restriction under
finance.accounts, and the shape check passes it.
Coverage and Ownership
- Role:
spoke - Owner family:
backend
Owned repo surfaces:
.agents/subagents/local_model_judge.md.agents/skills/puppy-one-harness/references/judging-contract.md
Non-owned surfaces:
consent-protocolhushh-webappbackend
Do Use
- Running the benchmark or the judge over a local model.
- Grading a review queue, or reviewing someone else's verdicts.
- Adding a model to the ladder, or reading the ledger for a trend.
- Grading the device's real daily jobs (the cron quality suite, see
.agents/skills/puppy-one-harness/references/cron-quality-suite.md).
Do Not Use
- Changing the PKM write path or the vault (
security-audit). - Provider routing or the on-device gate itself (
backend-runtime-governance). - Frontend surfaces for Puppy One (
frontend-surface-placement).
Read First
.agents/skills/puppy-one-harness/references/judging-contract.md.agents/subagents/local_model_judge.md
Workflow
- Collect outputs from the local model, then write a review queue. Controls are planted, shuffled per run, and unmarked; scoring reads them from the seal.
- Grade in a session that did not write the queue. This cannot be enforced from a script, so it is a discipline the report states rather than assumes.
- The answer key is the seal (rows, rules, seed) and
memory_judge_controlsin the harness. Never open or hunt for either; the manifest locates no control. - Every
wrongverdict cites the offending value verbatim. Ingest checks the citation against the output and discards it if absent, because an uncited failure is indistinguishable from a hallucinated one. - Grade every row. Ungraded rows void the run: skipping the hard ones would raise accuracy for free.
- Ingest, then append to the ledger. Record void runs too — dropping them makes the ledger an unbroken record of successes.
- Before comparing two runs, check the capability profile. A model tested via tool calling and one tested via JSON mode were not asked the same question.
Handoff Rules
- If the request is still broad or ambiguous, route it back to
backend. - If the finding is a provider-routing or gate issue, use
backend-runtime-governance. - If it touches the vault or the PKM write path, use
security-audit.
Required Checks
python3 .agents/skills/agent-orchestration-governance/scripts/agent_orchestration_check.py
python3 .agents/skills/codex-skill-authoring/scripts/skill_lint.pyDoes it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

